Government
Publications
Ontario
THE REPORT OF THE
Royal Commission on
Electric Power Planning
Chairman: Arthur Porter
VOLUME 2
The Electric Power System in Ontario
THE REPORT OF THE
Royal Commission on
Electric Power Planning
Chairman: Arthur Porter
February 1980
Published by the Royal Commission on Electric Power Planning
Printed by J.C. Thatcher, Queen’s Printer of Ontario
ISBN: The Report (9 volumes): 0-7743-4672-8
ISBN: Volume 2: 0-7743-4664-7
Design and production management: Ken Slater
Photocomposition: Shirley Berch
Text management and photocomposition facilities: Alphatext Limited
Graphics: Acorn Technical Art
Editors: T.C. Fairley & Associates; R.A. Grundy & Associates
For the Commission: Ann Dyer, Editorial Coordinator; DebbieAnne Chown, Terminal Operator
Royal Commission on Electric Power Planning
ARTHUR PORTER, Chairman
ROBERT E. E. COSTELLO,* Commissioner
GEORGE A. McCAGUE, Commissioner
SOLANGE PLOURDE-GAGNON, Commissioner
WILLIAM W.STEVENSON, Commissioner
ANN DYER, Executive Programme Co-ordinator
FREDERICK R. HUME, Q.C., Legal Counsel
PETER G. MUELLER, Senior Advisor
ROBERT G. ROSEH ART, ; Scientific Counsellor
ROBERT A. SCOTT, Q.C.,+ Legal Counsel
RONALDC.SMITH, Executive Director
*Resigned on 9 May 1977 due toill health
+ Resigned on 30 June 1977 to become Dean of University Schools, Lakehead University
{Lost his life in the air crash on September 4, 1976
Previous publications of the
Royal Commission on Electric Power Pianning
Shaping the Future. The first report by the Royal Commission on Electric Power Planning. Toronto,
1976
The Meetings in the North. Toronto, 1977
Outreach Guidebook. Toronto, 1976
Issue Paper 1: Nuclear Power in Ontario. Toronto, 1976
Issue Paper 2: The Demand for Electrical Power. Toronto, 1976
Issue Paper 3: Conventional and Alternate Generation Technology. Toronto, 1977
Issue Paper 4: Transmission and Distribution. Toronto, 1977
Issue Paper 5: Land Use. Toronto, 1977
Issue Paper 6: Financial and Economic Factors. Toronto, 1977
Issue Paper 7: The Total Electric Power System. Toronto, 1977
Issue Paper 8: The Decision-Making Framework and Public Participation. Toronto, 1977
Issue Paper 9: An Overview of the Major Issues. Toronto, 1977
A Race Against Time: Interim Report on Nuclear Power in Ontario. Toronto, 1978
Our Energy Options. Toronto, 1978
Report on the Need for Additional Bulk Power Facilities in Southwestern Ontario. Toronto, 1979
Report on the Need for Additional Bulk Power Facilities in Eastern Ontario. Toronto, 1979
The Report of the Royal Commission on Electric Power Planning
List of Volumes
The Report of the Royal Commission on Electric Power Planning is comprised of the following volumes:
Volume 1: Concepts, Conclusions, and Recommendations
Volume 2: The Electric Power System in Ontario
Volume 3: Factors Affecting the Demand for Electricity in Ontario
Volume 4: Energy Supply and Technology for Ontario
Volume 5: Economic Considerations in the Planning of Electric Power in Ontario
Volume 6: Environmental and Health Implications of Electric Energy in Ontario
Volume 7: The Socio-Economic and Land-Use Impacts of Electric Power in Ontario
Volume 8: Decision-Making, Regulation, and Public Participation: A Framework for Electric Power
Planning in Ontario for the 1980s
Volume 9: A Bibliography tothe Report
VOLUME 2
The Electric Power System in Ontario
Sushil Choudhury
The Author
SUSHIL CHOUDHURY was born in India and obtained a degree in electrical engineering there in
1972. He did graduate work in control systems at the University of New Brunswick where he obtained
an M.Sc.E. degree in electrical engineering in 1974. After that he obtained an M.A.Se. degree in
industrial engineering from the University of Toronto; his thesis for that degree concerned the long-
term planning of the electric supply system in Canada. In January 1977, he joined B.C. Hydro in
Vancouver, where he worked on system planning projects. In January 1978, he was seconded to the
Commission. Mr. Choudhury isa member of the Institute of Electrical and Electronics Engineers and is
a registered professional engineer in Ontario. He has published several technical papers in the areas of
control systems, power devices, and electric supply system planning.
Author’s Acknowledgements
The author wishes to acknowledge the contributions of Mr. K.H. Kidd of Leighton & Kidd Limited, Mr.
K.J. Slater of Slater Energy Consultants Inc., and Professors W. Janischewskyj and J.S. Rogers of the
University of Toronto. Chapter 5 is based entirely on a report done by Mr. Kidd for the Commission and
Chapter 6 was written by Mr. Slater. Professor Janischewskyj contributed the section on the planning
of bulk power transmission in Chapter 7. Dr. Rogers wrote the sections on the costs and benefits of
reliability in Chapter 4 and on the System Expansion Program Reassessment study in Chapter 7. Dr.
Rogers also provided many helpful suggestions throughout the writing of this volume. Thanks are also
due to Mr. Paul Burke, Mr. Richard Jennings and other members of the RCEPP research staff for their
useful comments.
Table of Contents
Foreword xi
EBrecutiveSummary xiii
Chapter One: Introduction 1
The Scope of This Volume; RCEPP Issue Paper No. 7; Outline of This Volume.
Chapter Two: The Electric Po wer Systemand Its Components 5
The Concept of a Power System; Power and Energy; The Components of an Electric Power System;
The Nature of the Demand for Electric Energy.
Chapter Three: Mix of Genera ting Resources 21
Comparative Costs of Generation Alternatives; Cost Comparison: Nuclear and Coal; The Reliabil-
ity and Performance of Various Technologies; The Operating Characteristics of Various Technolo-
gies; The Spectrum of Conventional Technologies; Fuel Requirements and Supply; Heavy-Water
Supply and Demand; Lead Time; System Considerations; Ontario Hydro’s Practice in Assessing
Generating Mix; Summary and Conclusions.
Chapter Four: The Reliability of the Electric Power System 39
Availability and Security; Reliability Evaluation; Evaluation of the Reliability of Supply to the
Customer; The Costs and Benefits of Reliability; Summary and Conclusions.
Chapter Five: Interconnections with Other Systems 59
Electricity Trade with Neighbouring Provinces and States; The Outlook for Electricity Trade; The
Role of the National Energy Board (NEB); “Electricity Exchanges” — A Canada-United States
Study; Interprovincial Interconnections; Interconnections within Ontario; Conclusions.
Chapter Six: Operation and Control of the System 69
The Operating Control of a Power System; Protection, Indication, Control, and Communications;
The Management of a Bulk Power System; Electrical Operation of the Power System; The Eco-
nomic Operation of the System.
Chapter Seven: The Planning of the Electric Power Systemin Ontario 77
The Changing Planning Environment; The General Approach to Planning Generation; Ontario
Hydro’s Long-Range Generation Forecasts; System Expansion Program Reassessment (SEPR)
Study; Ontario Hydro’s Current Generation Expansion Programme; Implications of Low Load
Growth; Planning Bulk Power Transmission.
Chapter Eight: The Impact of Alternative Technologies 97
Load Management; Electric Energy Storage; Co-Generation; Biomass and Refuse-Derived Fuels;
Summary and Conclusions.
Appendix A: Ontario Hydro’s Generating Resources 107
Definitions,
Appendix B: Summary of Characteristics of Conventional Generation Technologies 115
CANDU Nuclear Generation; Fossil-Steam Generation; Gas Turbine Generation: Hydroelectric
Generation.
Appendix C: Some Technical Consideration Related to Ontario’s Interconnections 123
NotestoChapters 125
Foreword
The Commission wishes to acknowledge the contributions to our Final Report made by the author of
this volume. The enormity of the task as well as the skill and tenacity with which it was performed are
testimony to the talents of Sushil Choudhury. Our work would have been immeasurably more difficult
without his assistance.
This volume, The Electric Power System in Ontario, focuses on a key issue area raised by the public
during the Commission’s public hearings process. The analysis, conclusions, and recommendations
reflect data received by the Commission in the form of public testimony and exhibits, consultants’
reports, and independent research and analysis by the author. We have relied heavily on this work in
formulating our own conclusions and recommendations in Volume 1. However, the views expressed in
this volume are ultimately the responsibility of the author. This document is therefore best viewed as a
background paper which attempts to draw together the detailed evidence and analysis available on this
complex subject, in a fashion which will be of use to the general public as well as to the technical
community.
The research and evolution of this document were directed and reviewed for the Commission by Philip
A. Lapp and Peter G. Mueller.
Arthur Porter, Chairman.
——————————— a ee
Foreword xi
ae me!
eT see
i — ,
TiN ‘fron [sa aU
rua wa os W ga cued nye % re a ee dmdly ae f imi AN,
- _ ‘ a - val : Pal a) 4
1, Relea ee ee S carmery “uivaget ay wy <<,
a4 a or ~ ‘oo ) a ae ’
. . a 4) 4 ' Zz 4 “et 2 = 1) A * =
4 |
7
; s
i
XN
Lie}
4
i Se as — ee a a ee ne ee ee ee roe
: . uy4 > 5 ‘
-. £ - é =
wh | ywasea i gre Pee "') . .
ee dt a am oo. | Bey i ° vy ae “=i are & I SPO | i
Executive Summary
The objective of this volume is twofold — to illustrate the concepts that are essential to the operation and
planning of a large electric power system, and to discuss the issues that are related to the technical
aspects of Ontario’s electric power system. As the first of the supporting volumes of the Commission’s
Report, this volume provides a comprehensive description of Ontario’s electric power system, with the
idea that this will make it easier to understand the economic, socio-environmental, and other broad
issues that are discussed in the subsequent volumes.
The technical aspects of Ontario’s electric power system are discussed in terms of the mix of generating
resources, system reliability, the interconnections with other systems, and the operation and control of
the system. Also discussed is the probable impact of alternate technologies on the planning of the
system by the end of this century.
The mix of generating resources in an electric power system is affected by many factors, important
among which are cost, reliability and performance, operating characteristics, fuel requirements and
supply, lead time, and total-system considerations. For base-load applications, anew CANDU nuclear
station has a significant economic advantage over a new fossil station fuelled either by U.S. coal or by
western Canadian coal. The economic comparison between coal and nuclear is not very sensitive to
changes in various parameters such as the cost of geological disposal of spent nuclear fuel, discount
rate, uranium prices, ete. Oil- and gas-fired plants are economic only for peaking and reserve. The
operating performance of the CANDU units indicates that they will be able to maintain base-load
capacity factors. While CANDU units are not suitable for load-following, they are capable of shut-down
on weekends and of operation at reduced output overnight.
Ontario Hydro’s fuel supply up to the end of this century appears to be reasonably secure. Its coal
requirements in the late 1990s are not expected to be much higher than the current consumption levels;
in the 1980s there will be a problem of over-supply of coal. No problem is foreseen in meeting the
peaking and reserve requirements of residual oil- and gas-fired stations. Hydro’s current uranium
contracts are adequate for the 80-year requirements of about 5,400 MW of uncommitted nuclear
capacity to be installed after the Darlington G:S. This is sufficient for capacity additions at least until
the mid-1990s. Although the lead time of a major generating facility could be reduced from 13 to eight
years by site-banking, there is no urgency about doing this. Sufficient time is available for public
participation and environmental assessment of any new site proposal by Hydro.
A major total-system consideration in the choice of a generating mix is the desirability of a diverse mix
to increase the system’s resilience and flexibility. Although economics is the major consideration in
Ontario Hydro’s planned generating mix, operating limitations and fuel diversity are also important.
Therefore, the planned long-term share of nuclear capacity is lower (at about 50 per cent) than that
suggested by economic considerations (about 60 per cent). The role of storage schemes, such as under-
ground pumped storage, in enhancing the operating flexibility of CANDU units when faced with
lower-than-expected load growth is limited, because their lead times are likely to equal those of a
nuclear plant.
The assessment of the reliability of an electric power system such as Ontario’s is extremely complex and
difficult. Studies that have been undertaken by Ontario Hydro indicate that Hydro is a front-runner
among electricity utilities in its efforts to determine a justifiable level of system reliability and thus to
advance the state of the art of reliability assessment. Hydro has developed a generation reliability
programme based on the frequency-and-duration-of-outages (F&D) method. The F&D technique is
considered to be far superior to the widely used loss-of-load-probability method. Hydro, along with its
neighbours, is also active in developing practical schemes for the quantitative evaluation of transmis-
sion reliability. Because of the complexity of these schemes, it will be some time before they find
widespread use among electricity utilities, Experience in Ontario indicates that the contribution of
failures in the bulk power system to interruptions of supply to customers has been small. Most interrup-
tions occur because of failures in the distribution system.
As aresult of recent studies undertaken by Ontario Hydro to match the costs and benefits of reliability,
Hydro has reduced its planned generation reserve from about 30 per cent to 25 per cent of the firm
peak. The reserve requirements, expressed as a percentage of the primary peak, are only about 17 per
cent, representing an 8 per cent reduction in capacity by load management. The combined effect of the
i se" ——“‘C‘CUW
Executive Summary — xiii
reduced reliability criterion and load management isa reduction in system capacity requirement in the
year 2000 by the equivalent of one station the size of Nanticoke.
Ontario’s interconnections with its neighbours (Quebec, Manitoba, Michigan, and New York) have
served their purpose well. They have demonstrated that their benefits far outweigh their costs. Strong
interconnections tend to increase a system’s resiliency. This and a number of other considerations
suggest the desirability of increasing electricity interchange, not only in absolute terms, but also
relative to total domestic demand. Ontario Hydro has a substantial surplus in generating capacity that
is expected to continue until the early 1990s. It is in Ontario’s interests for Hydro to make profitable
sales from this surplus, especially to its U.S. neighbours. The sheer size of the U.S. market, its heavy
reliance on oil-fired generation, and the delays encountered by U.S. utilities in commissioning coal-
fired and nuclear plants suggest a continuing market for Ontario’s surplus. Also, Hydro’s system is
larger and more diversified than those of any of the neighbouring U.S. utilities, which are mostly
investor-owned, profit-oriented, fragmented, and subject to capital constraints and taxation on
income.
Power-transfer capability with the U.S. is projected to decline sharply as internal transmission in
southern Ontario becomes fully dedicated to domestic loads. To take advantage of the export opportuni-
ties, construction of major new transmission facilities in eastern and southwestern Ontario is required.
Strengthening interconnections with Manitoba and Quebec will require comprehensive agreements
between the provincial and federal governments as well as between the electricity utilities. Such
interconnections will tend to accelerate the development of Canada’s remaining hydroelectric poten-
tial (for example, the Nelson River in Manitoba) and will take advantage of the increasingly comple-
mentary nature of the largely hydraulic systems in Quebec and Manitoba and the increasingly thermal
system in Ontario. A proper evaluation of firm hydraulic purchases from Quebec and Manitoba can only
be made ina broad provincial total-energy framework.
The strengthening of the transmission system in eastern Ontario, especially for the supply to Ottawa,
is urgently needed. This will also facilitate electricity interchange with Quebec and New York. In
southwestern Ontario, the objectives of incorporating Bruce B, supplying future loads, and strength-
ening the transmission network and the interchange capability with the U.S. can best be met, from a
purely technical viewpoint, by constructing a 500 kV line from Bruce to London. However, there are
socio-environmental and land-use implications of such an alternative that must be fully investigated.
The emergent technologies that are expected to have the maximum direct impact on Ontario’s electric
power system before the end of the century are load management, electric energy storage, co-genera-
tion, and generation from biomass- and refuse-derived fuels. Ontario Hydro’s participation in load
management with the municipal utilities over the next few years is worthy of support because it will
provide the needed assessment of the cost and the public acceptance of load management. Hydro’s
target for managed load by 1992 is 1,800 MW. This target will have to be reassessed in the light of the
experience gained in the 1980s with load management and in the light of other ways of meeting the
peak load, such as peaking hydraulic and storage. Among the various alternatives for large-scale
storage, underground pumped storage appears to be the most economical. The economic justification
for developing storage will depend on the need for new peaking capacity and the availability of surplus
nuclear energy for pumping. Hydro’s system could accept from 0 to 2,000 M W of pumped storage by the
late 1990s, based on an average annual rate of load growth of from 3 to 4.5 per cent.
The growth of co-generation in Ontario will be influenced greatly by the nature of the financing, the
price of boiler fuels, and the price of electricity purchased from Ontario Hydro. Any growth in indus-
trial co-generation will tend to replace Ontario Hydro’s base-load requirements. Parallel operation of
many small co-generators with the Ontario Hydro system will increase the complexity of overall system
operation. However, many small co-generators will tend to reduce overall system reserve requirements
and the stand-by charges faced by the co-generators. Co-generation has the added advantages of more
efficient utilization of fossil fuels, diversification, and decentralization.
Central wood-fuelled generation is not economic at present in comparison with coal-fired and CANDU
nuclear generation. Although the municipal refuse generated in Ontario has a potential to produce
about 3.5 per cent of Ontario’s demand for electricity, it is more likely to be used both to generate
electricity and provide district heat. Utilizing the energy potential of refuse is one of the best ways to
solve the waste-disposal problem and provide some diversity and security in the fuel base, by making
use of an indigneous, renewable, and cheap source of energy.
xiv. The Electric Power System
CHAPTER ONE
Introduction
The electric power system in Ontario is one of the largest and most sophisticated power systems in the
world. It has been supplying the growing demand for electricity in the province with a high degree of
reliability and at reasonable cost since the early part of this century. The system is for the most part
identifiable with Ontario Hydro, which supplies more than 90 per cent of the province’s demand for
electric energy. The remainder is either supplied by privately owned utilities, notably the Great Lakes
Power Corporation and the Canadian Niagara Power Company, or generated by industrial establish-
ments for their own use. In 1978, Ontario Hydro supplied approximately 90 billion k W-h to 334 munici-
pal utilities, 785,000 rural customers, and 100 large industrial users. Its power system comprises ap-
proximately 23,000 MW of generating capacity, based on a variety of primary energy sources, and
32,000 km of transmission and distribution lines of various voltages, serving Ontario’s population of
eight million. For planning and administrative purposes, Ontario Hydro’s system is conveniently
divided into two systems which are connected by a single tie-line. The East System, by far the larger
and serving the geographic regions of southern, eastern, and northeastern Ontario, is an integrated
grid system, while the West System, serving the northwestern regions of the province, is essentially a
linear system. The dividing line between the two systems runs north-south through the community of
Wawa, north of Sault Ste. Marie. However, electrically, the two systems operate in synchronism as one
system.
The Scope of This Volume
The objective of this volume is twofold. First, to illustrate the basic concepts that are vital to the
planning and operation of an electric power system, and, second, to discuss the issues that are related to
the technical aspects of electric power planning in Ontario. As the first of the supporting volumes of the
Commission’s Final Report, this volume provides a comprehensive description of Ontario’s electric
power system, in the hope that the economic, socio-environmental, public-participation, and other
issues discussed in the subsequent volumes may then be better understood. For example, the issue of the
use of land for transmission lines can be seen in perspective only after recognition of the role of
transmission lines in a total power system, and the public’s involvement in the decision-making process
must be assessed in the light of the complexity of an electric power system.
Most of this volume is devoted to the technical considerations relating to the electric power system, such
as reliability, operation and control, and fuel requirements, because these are the first to affect the
design of a system. The broader social, economic, environmental, and political considerations are the
subjects of other volumes. Since it is impossible, and indeed impractical, to isolate them from technical
considerations, some overlap is inevitable.
The majority of the issues the Commission faced, related to bulk power transmission planning in
Ontario, were socio-economic or socio-environmental. Therefore, bulk power transmission is discussed
here to the extent that it is vital toan understanding of the overall design of the system. For example, in
any consideration of system design, the security of bulk power transmission is pivotal. Among the
generation alternatives analysed in detail in this volume are the ones that are conventionally used in
the province and have the potential for large-scale development in the 1980s and 1990s. New and
emergent generation technologies are described in Volume 4. However, some of them, with a potential
to affect system planning and operation in this century, are examined in this volume.
RCEPP Issue Paper No. 7
Many of the issues that are discussed in this volume were raised during the Commission’s public
information hearings (April 1976 to January 1977) and subsequently summarized in the Commission’s
Issue Paper No. 7, entitled ‘'The Total Electric Power System”, published in April 1977. The issues were
listed according to four characteristics of a total power system, identified by the Commission as reliabil-
ity, operational concepts, the mix of generating stations, and interconnections. The following is an
outline of that issue paper and of the issues that are considered in this volume.
a ee ee ee ee
Introduction 1
Reliability
A high level of reliability isa most desirable feature of an electric power system. A factor that enhances
reliability is a total-system configuration that is flexible and resilient. A multiplicity of generating
stations rather than a single massive one and a network of transmission and distribution systems
rather than asingle ultra-high-capacity link make a system less vulnerable to the effects of component
breakdowns and outages. Other factors that enhance system reliability are an adequate generating
reserve, a secure bulk power transmission and distribution system, optimal system operation and
control, and interconnections with neighbouring systems.
Some issues that are relevant to reliability are: What is the impact of various planning factors on
system reliability? In order to balance the costs of high levels of reliability against their benefits, to
what extent could the generation reserve requirements be reduced? What is the role of load manage-
ment in reserve reductions? Are Ontario Hydro’s reliability evaluation and assessment techniques
adequate? These issues are discussed in Chapters 4, 7,and8.
Operational Concepts
Because of varying demand patterns, generating units are operated in one or more of the following
four modes: base load, intermediate load, peaking, and reserve. The cost economics of the resources
available in Ontario make nuclear plants suitable for base load, coal-fired plants suitable for base and
intermediate loads, and oil- and gas-fired units suitable for peaking and reserve. Hydraulic plants are
well suited for any of these modes but this versatility is constrained by the limitations on water availa-
bility. Reliable and economic operation of a system such as Ontario Hydro’s requires a sophisticated,
computerized information monitoring and control system. Some of the issues concerning operational
aspects are: the potential for load management and energy storage for peak shaving, the operational
implications of centralized and decentralized generation, and the impact of alternate generation
sources such as industrial co-generation. The operational concepts and issues are discussed in Chapters
6and8.
The Mix of Generating Stations
A diverse mix of components is often desired in any complex physical system so that a catastrophic
breakdown of asingle major component does not necessarily lead to the failure of the total system. This
also applies to the mix of generating resources for an electric power system. In Ontario, the existing
mix includes hydroelectric, nuclear, and coal-, oil-, and gas-fired stations. In 1978, hydroelectricity
supplied about one-third, coal and nuclear generation approximately one-fourth each, and gas, oil, and
purchases the rest of the electric energy demand. The choice of a particular mix is determined by
factors such as cost, efficiency, the security of fuel supplies, operating limitations, environmental con-
sideration, and the availability of suitable sites. Some of the related issues are: the cost-competitiveness
of nuclear and coal-fired stations; the reliability and performance of Ontario Hydro’s large thermal
units; the future availability of fuels and heavy water; the lead times of large centralized generation
facilities; and the impact of load management, energy storage, and co-generation on future generat-
ing mix. These issues are discussed in Chapters 8, 7,and8.
Interconnections
The purpose of interconnections among electric power systems is to exchange power in emergencies or
for reasons of economy. Ontario Hydro is linked with Hydro-Québec and Manitoba Hydro in Canada
and with the New York and the Michigan Power Pools in the U.S. Interconnections enhance reliability
and reduce reserve requirements by facilitating emergency exchanges of power and by taking advan-
tage of the seasonal and time-zone diversity in the loads of the neighbouring systems. Interconnections
may also reduce or delay the need for additional generating capacity, through arrangements for firm
power imports. Some of the major issues raised by interconnections are: the potential for export of
Ontario Hydro’s surplus generating capacity in the 1980s; the implications of reliance on imports of
hydroelectric power from Quebec and Manitoba in the 1990s; the implications of strengthening On-
tario Hydro’s interconnections with its neighbours; and, the role of intra-provincial interconnections.
Chapter 5 discusses these issues.
2 The Electric Power System
Outline of This Volume
This volume comprises eight chapters supplemented by three appendices. Chapter 2 introduces the
concept of an interconnected power system and its major components — generation, bulk power trans-
mission, distribution, and end use. It does this by tracing the journey of electricity from its production
from primary energy sources to its ultimate disposal. The various generation and transmission techno-
logies that are at present available are described briefly. Also illustrated are the concepts of load factor,
capacity factor, and load-duration curve, which are central to the planning and operation of a power
system.
In Chapters 3, 4, 5, and 6, the four characteristics of the electric power system in Ontario that were
identified in RCEPP Issue Paper No. 7 are discussed. These are generating mix, reliability, intercon-
nections, and operation and control. Chapter 3 illustrates the important distinction between the mix of
capacity and the mix of energy resources and provides a comparative analysis of factors associated
with various technologies that affect the choice of a mix in Ontario. Among the factors considered are
costs, reliability, operating characteristics, fuel requirements and supply, lead time, and total-system
considerations such as resiliency and flexibility.
Chapter 4 considers the reliability of an electric power system. Starting with the traditional ap-
proaches to reliability evaluation, such as availability and security, this chapter underlines the diffi-
culty of assessing a desired level of reliability, because of the complex nature of a power system. The
effect on reliability of such factors as the outage rate of a generating unit, unit size, and interconnec-
tions with other systems is explained. Finally, the chapter describes the efforts of Ontario Hydro to
advance the state of the art in reliability evaluation and assessment.
Chapter 5 deals with the interconnections between Ontario Hydro and its neighbours. After comment-
ing on the outlook for electricity trade in the light of Hydro’s projected surpluses, it reviews the
conclusions and recommendations of two recent studies, dealing with the Canada-U.S. and interpro-
vincial interconnections, respectively. Finally, a brief description of the interconnections between
Ontario Hydro and other smaller members of the province’s electric power system is included to under-
line the role of those members in Ontario’s electric power planning.
In Chapter 6, the basic concepts that govern the operation and control of an electric power system are
presented. The functions and responsibilities of the various levels of management and of the opera-
tional control hierarchy for a smooth day-to-day operation of the total system are discussed. The dis-
tinction between the electrical operation and the economic operation of a power system is explained.
Chapter 7 analyses the long-range planning considerations used by Ontario Hydro in the past and
summarizes recent developments affecting these considerations, the utility’s current expansion plans,
and implications of a low-growth scenario. The chapter considers the methodology of Hydro’s System
Expansion Program Reassessment Study and the general nature of the results. Also presented are
some technical considerations relating to the future of bulk power transmission in Ontario.
Finally, in Chapter 8, the impact of some potential alternate technologies on the operation and plan-
ning of the electric power system in Ontario is assessed. Among the technologies considered are load
management, energy storage, co-generation, and biomass and refuse-derived fuels.
Appendix A gives a station-by-station listing of Ontario Hydro’s generating resources. Appendix B
outlines the characteristics of the conventional generating technologies in Ontario that affect the
choice of a generating mix, and Appendix C discusses some technical considerations that relate to
Ontario Hydro’s interconnections with its neighbours.
SSS SSS SSS a a a
Introduction 38
a
ae
VIN a wh nine
8
mm |
+e
_ pest <aly
i,
Jes Dbl ne
ree yer ie
ry die eae
; - ' re
eer pan alia} ane sais
iin ence ~~ e ree “
-
é
= me : ae | 7
Te ” ss as oul wm et ee pnts
ee af ¥ = > os i 7 / ?
, \ 9s a = $ * eo
; “ 4 oe
fea » io Y ; > 7
. 248
’ “4 : wy
S § ao oe
|
:
Ph 7
:
“ee 7
oe
A
: . ek
j j ‘fay Ot .e.7' au" ie One
¢ ‘ we a : ; At 1 5
7 : Ty ; i Ws 0 rif 43 SPP ve
a 7
4
Fig. 2.1. p. 15
CHAPTER TWO
The Electric Power System and Its Components
The Concept of a Power System
Electric energy does not occur in nature in a form in which we can use it. It must be converted from a
primary source of energy such as falling water, coal, uranium, oil, or natural gas. The process of
conversion is called generation. After being generated, electric energy flows over miles of bulk power
transmission and distribution lines, and undergoes a series of voltage transformations before it is
ultimately used in our homes, industries, and commercial buildings. The utilization of electric energy at
a point of use, or for a type of use, is generally referred to as “electrical load”. An electric power system
may be defined as a number of generating stations and a multitude of customer loads interconnected
through a network of transmission lines, transformer stations, and other transmission and distri-
bution installations. The essential quality that makes this collection of apparatus a power system is that
every one of its generating stations contributes to the supplying of every customer load.
The existence of large power systems may be explained by two qualities that customers value in their
supply — reliability and economy. Power systems enhance the reliability of supply to the consumer. All
power plants have to be taken out of service from time to time, either because of unscheduled shut-
downs or for scheduled maintenance. If each load centre were supplied by a single nearby generating
plant, few bulk power transmission lines would be needed, but there would be a complete local black-out
every time a plant had to be shut down. Without a bulk power transmission network, the only way to
avoid this would be to duplicate the plant at each load centre. This would be costly and it would still not
guarantee a continuous supply of electricity, because the two plants might have trouble simultaneously,
either by coincidence or by common cause. With a bulk power transmission network that is fed by
several generating stations, much greater reliability of the total system can be realized with a much
smaller reserve of generating capacity. Table 2.1 illustrates this concept. It considers 10 customer
loads, each of 10 MW. In case 1, each load is supplied by its own isolated generators and the probability
of generation deficiency with a 100 per cent reserve capacity is 1 per cent. In case 2, however, when all
loads and generators are interconnected, the probability of generation deficiency with a generating
reserve of only 50 per cent drops to 0.225 per cent.
Table 2.1 Illustration of the Improvement in Reliability in a “Power System’’
Case 1: Individual Generator - Load Arrangement
Load: 10 MW
Generation: 2 X 10 MW units, each with a 10 per cent probability of failure
Reserve: 100 per cent
Probability of not meeting load: 1 per cent
Reliability: 99 per cent
Case 2: Power System
Load: 100 MW
Generation: 15 > 10 MW units, each with a 10 per cent probability of failure
Reserve: 50 per cent
Probability of not meeting load: 0.225 per cent
Reliability: 99.775 per cent
Source: RCEPP.
Power systems also provide considerable economy over the alternative arrangement, in which single
generating stations serve only certain dedicated loads. The nature of individual loads is such that their
peaks do not occur at the same time. This diversity causes the aggregated load to have a peak that is
lower than the sum of the peaks of the individual loads, as illustrated in Figure 2.1. This allows a power
system to have less overall generating capacity than the sum of all the individual generating capacities
that would be required to serve individual loads separately. Also, because the overall load is greater ina
power system than it is for individual customers, the power system can incorporate larger generating
units, offering economies of scale. Power systems also permit significant cost savings in the operation of
the system. Ontario Hydro’s system has various kinds of generating plants — hydroelectric; coal, oil,
The Electric Power System and Its Components 5
gas, and uranium steam-electric; and combustion turbine plants. Hydroelectric plants have no fuel
costs, and nuclear plants generate electricity at substantially lower fuel costs than fossil-fuelled plants.
During the off-peak periods, great cost savings can be made by supplying the entire system from such
stations. This can only be done with an extensive bulk power transmission network.
Power and Energy
The term electricity is quite often used as a substitute for both “electric power” and “electric energy”.
However, the distinction between power and energy must be clearly understood. Engineers and scien-
tists use “power” when referring to the rate of flow of “energy”. For example, an electric heater left
half-on for two hours will heat a well-insulated room to the same temperature as the same heater left
full-on for one hour. This means that the amount of electric energy consumed is the same in both cases,
but in the second case the power level, i.e., the rate at which the electric energy is being consumed, is
doubled.
The relationship between power and energy is well demonstrated by the electricity meter in your home.
Inside this meter is a large thin circular disc that rotates and drives a series of registers that count its
rotations. The disc rotates slowly when a small amount of electricity is being used, and more rapidly
when the demand is heavier. The rate of rotation of the disc is a measure of the rate at which electric
energy is being used, i.e., of the level of power being supplied. The total number of rotations of the disc
in a given period is a measure of the electric energy that has been consumed. Thus, power is measured at
an instant in time; a unit of power does not have the dimension of time in it. In contrast, the measure-
ment of energy must be done over a period of time — a minute, an hour, or a year.
Electric power is often measured in kilowatts (kW) and electric energy in kilowatt hours (kW-h), a
kilowatt hour being the amount of electric energy consumed when a device of 1 kW power rating is
switched on for one hour, or one of 0.5 k W for two hours, or one of 10 k W for six minutes. Since one of the
most commonly available primary sources that is used to generate electric energy is thermal energy, it
is useful to consider the relationship between the kilowatt hour and commonly used units of heat such as
the British Thermal Unit (BTU) and the kilocalorie (kcal).
When a kilowatt hour of electric energy is converted into heat at 100 per cent efficiency, e.g., in
resistance heating, it produces 3,412 BTU or 860 kcal of heat. However, the number of British Thermal
Units required to produce 1k W-h of electric energy is significantly higher than 3,412, due tothe limited
conversion efficiency of heat engines. For example, a large modern coal-fired steam-thermal generat-
ing station has an efficiency of about 38 per cent, which means that, to produce 1 kW-h of electric
energy, approximately (3,412 x 100/38 =) 9,000 BTU of heat energy in coal is required. The kilowatt
hour is also a unit of mechanical work, but it isnot commonly used for that. The basic unit of work is the
joule, which is equivalent toa watt second. Thus 1k W-his equivalent to 3.6 megajoules (MJ).
Since, at present, electric energy cannot be stored economically in large quantities, the concept of
power is important for electric power systems. It is related to the capability of a system to meet the
customers’ demand for electric energy at the power level they desire and when they desire it. Power
may be used at a certain level only infrequently, and then only for short periods. The electric supply
system (generation, transmission, and distribution) and the wiring in any house must be designed to
handle the maximum power level required or specified.
Power and energy are also related through the cost economics of the various generation technologies
that are available. The capital cost of a plant is related to its capacity, or maximum power capability. The
cost of energy production is, essentially, the cost of the fuel used. Thus for power demands of very short
duration (i.e., with a small energy demand), the plants with the lowest capital cost are used. Combustion
turbines fall into this category and are quite commonly used for short-duration applications. For power
demands that occur virtually continuously (i.e., with a large energy demand), on the other hand, plant
selection criteria emphasize lowest total cost (fuel as well as capital). High-capital-cost plants may
satisfy this requirement if their fuel costs are quite low. In Ontario, hydraulic and nuclear plants fall
into this category. The methods of cost minimization for power demand levels of various durations will
be explained in Chapter 3.
6 The Electric Power System
Fig. 2.2: p. 16
Pec oopo ul
The Components of an Electric Power System
The journey of electricity in a power system, from its production from primary fuels to its ultimate
disposal, comprises four major steps — generation, high-voltage bulk power transmission, distribution,
and end use. The components of an electric power system will be considered under those four headings.
A schematic representation of Ontario Hydro’s power system is provided in Figure 2.2. The “area
supply” and “subtransmission” stages shown in Figure 2.2, the two additional voltage transformation
steps between generation and end use, are indicative of the size and spread of Ontario Hydro’s system.
The Generation of Electric Power
The most commonly used techniques for the generation of electricity are hydroelectric, steam-thermal
electric, and combustion turbine. In a typical hydroelectric generating plant, rivers fed by rain or
melting snow are dammed to form a reservoir. When the stored water is allowed to flow from the
surface of the reservoir through a hydraulic turbine at a lower elevation, the potential energy in the
water is converted into kinetic energy, by falling, and then into mechanical energy, by the turbine. An
electricity generator connected to the turbine shaft rotates and converts this mechanical energy into
electric energy. The water continues its downward flow, to other reservoirs or directly into the lakes or
oceans.
Hydroelectric plants have some important advantages over thermal plants. The fuel, that is, the grav-
itational potential energy supplied by the sun, is free and renewable. Hydroelectric conversion is a one-
step conversion of mechanical energy to electric energy with greater than 90 per cent efficiency.
Further, there is no emission of pollutants and no need for cooling water. Hydroelectric plants also have
considerably longer lives (about 70 years) than thermal plants (about 40 years) and have much higher
reliability. However, they may suffer from the disadvantage of having an adverse impact on the local
environment, thatis, ecology, aquatic life, and scenic beauty.
A variant of the hydroelectric generation concept, called pumped storage, is the only practical method
at present for large-scale storage of electric energy. Pumped storage plants have been made possible by
the development of highly efficient reversible pump-turbines and motor-generators. During off-peak
hours, surplus electricity from low-fuel-cost plants is used to pump water from a low elevation to a
storage reservoir above the station. When the demand increases, this water is released and used to
generate electricity, as a substitute for some much more expensive form of peaking generation. The
overall efficiency of a pumping-generating cycle is about 70 per cent.
In steam-thermal electricity generation, electricity is generated from the stored energy in a fuel such
as coal, oil, natural gas, or uranium, through the production of steam. By the fission of uranium-235 ina
nuclear reactor, or by the burning of fossil fuel under a boiler, heat energy is produced, and this heat
energy is used to convert water into high-pressure and high-temperature steam. The steam is ex-
panded in a series of steam turbines, causing the turbines, and an electricity generator connected to
their common shaft, to rotate and produce electricity. By the time the steam leaves the turbines, it has
been considerably reduced in temperature and pressure, but it is still quite hot. A heat exchanger
condenses the steam back into water by transferring much of its heat to a body of cold water called
cooling water. The condensate is then pumped back intothe boiler to start the next cycle.
Figure 2.3 shows the steam cycle for a modern fossil-fuelled generating station. The thermal efficiency
of a steam cycle is defined as the ratio of useful energy output, i.e., electric energy, to thermal energy
input. Figure 2.3 shows that the efficiency of a fossil-fuelled station is about 37 per cent, 63 per cent of
the energy in the fuel being dissipated as ‘‘waste” heat to the atmosphere, to cooling water, and to other
heat sinks.
The steam cycle in a nuclear power plant is also explained by Figure 2.3. Instead of a boiler, there is a
nuclear reactor in which uranium is “burned”. The heat of the fission reaction is transferred to water to
convert it into steam, either directly, as in a boiling-water reactor (BWR), or through a heat transport
liquid such as heavy water, as in a CANDU reactor. There is little heat loss to the atmosphere in a
nuclear plant. Also, the steam entering the turbine is at a considerably lower temperature in a nuclear
power plant than in a fossil-steam plant. Ina CANDU reactor, the steam temperature is about 250°C,
compared with about 540°C in a fossil-steam plant. The lower steam temperatures of nuclear plants
limit their thermal efficiency to about 30 per cent. Since only a negligible amount of the waste heat is
dissipated to the atmosphere, the remaining 70 per cent must be discharged to the cooling water.
In a combustion turbine cycle, air is compressed, and this high-pressure air is mixed with fuel and fed
The Electric Power System and Its Components 7
into a combustion chamber, where it is ignited. The resulting high-temperature gaseous combustion
products are expanded in a gas turbine, causing it to rotate and drive an electricity generator. The air
compressor is also driven by the turbine. In a simple cycle, the gas, after expansion in the turbine
(although still quite hot), is discharged into the atmosphere. In a heat-recovery cycle, the exhaust gas is
sent back to “preheat” the incoming air-fuel mixture before being discharged into the atmosphere.
The efficiencies of the simple and heat-recovery cycles are 27 per cent and 33 per cent, respectively.
Since the moving parts of a gas turbine are exposed directly to the combustion products, fuels with
corrosive properties should not be used. The most commonly used fuels are distillate oil and natural gas.
Ontario Hydro uses No. 2 distillate oil in its combustion turbine plants.
When the heat in the hot exhaust gases from a gas turbine plant is transferred to a steam boiler to raise
steam for use in electricity generation, such an arrangement is called a combined-cycle plant. Interest
in such plants has grown in recent years because of their potential for efficiencies higher than those of
the large fossil-steam plants. The use of exhaust gases from gas turbines for steam production is,
however, not new. In the southwestern United States, the petrochemical industry utilizes this concept
to help satisfy the demand for electric energy as well as for process-steam. The General Electric Com-
pany and the Westinghouse Corporation are the leading exponents of the combined cycle in North
America. Many utilities in the U.S. have combined-cycle capacity, either installed or-under construc-
tion. Operating experience with these installations will have a significant effect on the extent to which
combined-cycle plants will be used in future utility systems.
In Ontario, Ontario Hydro supplies more than 90 per cent of the provincial demand for electric energy.
The remainder is provided by a number of small private utilities, for example, the Great Lakes Power
Corporation, the Canadian Niagara Power Company, and the many industrial establishments that
have hydraulic and thermal generating facilities for their own use. Many of the industrial thermal
plants are used in a co-generation mode to supply process-steam as well as electricity.
Until 1951, Ontario Hydro’s generating system was based exclusively on hydroelectric stations. In
195], the first two coal-fired thermal-electric generating stations, the 1,200 MW Richard L. Hearn
Generating Station (GS) on the Toronto waterfront and the 264 MW J. Clark Keith GS at Windsor,
were put in service. Since then, Ontario Hydro has added four more coal-fired stations and one residual
oil-fired station to its system, bringing its total fossil-fuelled capacity to about 12,000 MW. Hydro
began its nuclear power programme in 1962 with the incorporation of the 22 MW Nuclear Power
Demonstration GS of the CANDU design into the provincial power grid. In 1967, the 200 MW Douglas
Point Nuclear GS on the shore of Lake Huron was brought on line. Hydro’s first large-scale CANDU
plant is the 2,000 MW Pickering GS outside Toronto which began feeding power into the provincial
grid in 1971. A 3,000 MW CANDU nuclear station was recently added to the system at the Bruce
Nuclear Power Development complex near the Douglas Point GS.
Appendix A provides a station-by-station listing of all of Ontario Hydro’s generating resources, exist-
ing, under construction, and committed. The present system represents a balanced mix of the three
primary energy sources that are available to the province — water power, uranium, and fossil fuels. In
1978, these sources each supplied roughly one-third of the provincial electric energy demand. Coal,
imported from the U.S., was responsible for most of the electricity generated from fossil fuels.
All of Ontario Hydro’s steam-thermal generating stations are located along the shores of the Great
Lakes. As already noted, between 50 and 70 per cent of all the heat produced in a thermal power plant is
rejected to cooling water. The Great Lakes provide a relatively inexpensive source of cooling water for
this purpose. The cooling of Hydro’s generating stations is based on the “once-through” cycle; cold
water is drawn from the lake, circulated through steam condensers, and returned to the lake at a
higher temperature. The difference in temperature between the incoming and the outgoing water isin
the order of 10°C. The cooling of a 3,400 MW nuclear plant would require about 10 million litres of
cooling water per minute. That is to say that each kilowatt hour of electric energy generated requires
about 180 litres of cooling water.
Other methods of cooling are cooling towers and cooling ponds. In cooling towers, the warm water from
thermal station condensers is circulated and brought into contact with the ambient air, to accomplish
the heat exchange. Cooling towers are quite common in the U.S. and in Europe. Cooling towers tend to
be monstrous in size — in the order of 400 feet high with an equally large base diameter — and about
twice as expensive to build and operate as once-through cooling systems.
Cooling ponds are similar in concept to cooling systems using natural lakes or reservoirs except that
Se a aa Oe ee ae Re Ne ee sone Ree ee a Se
8 The Electric Power System
they tend to be smaller and they are sometimes man-made. Warm water is discharged into the pond, it
spreads over the surface and is cooled, largely by evaporation, and then it is returned to the plant.
Because of their low heat-transfer rates, cooling ponds require a considerable area of land — approx-
imately one to two acres per megawatt of generating capacity. The size of the pond may be reduced
drastically (by a factor of 20) if provision is made for spraying the discharge water into the air to
increase evaporation. The disadvantages of spray cooling are the cost of the power that is required to
operate the sprayers (about 1 per cent of station capacity) and the reduction in the station’s efficiency
that results from permitting the discharge of water ata higher temperature.
The issues associated with the environmental impacts of electric power generation in general are
discussed in detail in Volume 6 of this Report.
Bulk Power Transmission
After electric power has been generated, the next stage of its journey to the ultimate user is the bulk
power transmission system, which delivers power from the generating stations to the receiving termi-
nal stations. From the receiving terminal stations, power is carried to area-supply transformer sta-
tions, and from there to the familiar local transformers and thence to the users. Ontario Hydro’s bulk
power transmission system is predominantly at 230 kV. Recently constructed lines operate mainly at
500 kV, and there are a few old ones at 115 kV. The transmission lines that were in operation in the
Ontario Hydrosystem at the end of 1977 are shown in Table 2.2
Table 2.2 Ontario Hydro's Transmission Lines and Circuits, as of December 31, 1977
Line voltage Circuit length Tower-line length
(kV) (km) (km)
500 1,038 1,038
345 5 5
2306 12,779 9,002
115» 11,409 8,882
Total length Boreal 18,927
Notes:
a) The need to distinguish between circuit length and tower-line length arises from the fact that some lines have multiple circuits (in the case of Ontario Hydro,
almost all of them are double-circuit). Thus, a 100 km double-circuit line Segment is counted as 100 tower-line kilometres, but as 200 circuit kilometres.
b) The majority of the 115 kV circuits and some of the 230 kV circuits are actually part of Ontario Hydro’s area supply system. The exact lengths under the bulk
transmission category were not available.
Source: Ontario Hydro Statistical Yearbook, 1977.
The development of bulk power transmission in Ontario began in 1910 witha 115 kV system between
the hydroelectric plants at Niagara Fallsand major load centres in the Toronto-Hamilton area. In 1928,
230 kV was introduced, and the transmission system was expanded mainly at this voltage until the mid
1960s when Ontario Hydro installed its first 500 k V transmission line, from Moose River to Toronto.
Until the early 1950s, Ontario Hydro’s generating stations were predominantly hydroelectric, many of
them located at some distance from the load centres, and thus, long transmission lines were considered
inevitable. The introduction of steam-thermal power plants, fuelled first by fossil fuels and then by
uranium, changed the situation, because the siting of such plants is much more flexible. Indeed, the
coal-fired Hearn and Lakeview plants were built on the outskirts of Toronto. It soon became clear,
however, that the flexibility in locating thermal plants to optimize transmission line layouts was con-
strained by such factors as air pollution of urban areas (especially by coal-fired plants), public concern
about the risk of accidents at nuclear plants, the need for great quantities of cooling water, and the
realization that extensive economies could be achieved by grouping several large generating units at
one site.
These constraints and considerations led Ontario Hydro to build its generating stations close to but
outside major load centres on the shores of the Great Lakes. The need for a bulk power transmission
grid to interconnect these generating stations with the load centres was accepted because of the
resulting improved system reliability and reduced operating costs, as already explained.
The growth in bulk power transmission voltage in Ontario was a result of the growth in power trans-
mission requirements. Although the voltages at which power is generated in large modern generating
a
The Electric Power System and Its Components 9
units are often lower than 25 k V, transmission voltages are considerably higher (230 k V or 500 kV), in
order to limit transmission losses. The power lost in a line as heat is proportional to the square of the
current (if the current is doubled, the heat loss is quadrupled). Since the power transmitted is propor-
tional to the product of the voltage and the current, the best way to reduce the loss is to transmit power
at low levels of current and correspondingly high voltages.
In the early days of power system development, electricity was generated, transmitted, and used as
direct current (DC). DC voltages cannot readily be transformed, i.e., changed from one level to another.
Since the use of DC at high voltages was inconvenient, if not impossible, DC generation and transmis-
sion was restricted to low voltages, which are extremely uneconomical for long distances or for large
amounts of power. Alternating current (AC), due to the nature of the magnetic fields associated with it,
can be transformed efficiently to any desired level, and thus can be transmitted at high voltages and
used at convenient lower levels. AC generation, transmission, and distribution replaced DC, and most
power systems in the world today are AC. The highest AC voltage level used in Canada at present is 735
kV, on Hydro-Québec’s system. Voltages in the 345 kV to 765 kV range are called extra-high voltages
(EHV). Voltages higher than 765 kV, e.g., 1,000 k V and 1,500 kV, are called ultra-high voltages (UHV).
765 kV is the highest voltage used on commercial installations anywhere in the world today, but in the
next decade or so UH V transmission may be installed on some systems.
The choice of bulk power transmission voltage is determined by the amount of power to be transmitted,
the distance of transmission, and the configuration of the system. The efficiency and economics of
transmission are the criteria for the selection of a particular voltage level. Table 2.3 illustrates the
capital costs and the right of way requirements for transmitting 4,000 MW over a distance of 160 kmat
various transmission voltages. It is clear from Table 2.8 that, for this particular application, 500 kV is
an appropriate choice. Because it is Ontario Hydro’s policy to locate large thermal stations close to load
centres, it has selected 500 k V for the future bulk power transmission network.
Table 2.3 Capital Costs and Right-of-Way Requirements to Transmit 4,000 MW over 160 km
Transmission voltage (kV) 230 500 765
Typical circuit capability (MW) 280 1,920 4500
Number of circuits 14 2 l
Capital cost (1976 dollars) 200 x 10° 60 x 10° 70 x 106
Right-of-way width (m) 285 75 93
Land use (hectares) 4,560 1,200 1,488
Sources: Ontario Hydro, “Transmission-Technical’’, submission to the RCEPP, March 1976, Exhibit 5. Ontario Hydro, “Transmission Planning Processes’’,
submission to the RCEPP, June 1976, Exhibit 22.
The power capability of a transmission line varies with the length of the line and its current-carrying
capacity, as well as with the current-carrying capacity of other transmission equipment such as trans-
formers and circuit-breakers. The power capability decreases as the distance increases. The design of a
given transmission line is greatly affected by the configuration of the total bulk power system and by
the associated stability and security criteria. Both aspects of transmission will be considered in Chapter
4. What is meant by stability and security will be explained briefly here.
Allof the generating units (that is, synchronous alternators) on the Ontario Hydro system, and on all of
the systems with which it is interconnected, operate in synchronism, rotating at speeds that are in lock-
step with the North American standard frequency of 60 cycles per second. It is essential that all the
generators stay in lock-step. If generators go out of synchronism, violent swings in generator and
transmission loadings occur, which could, quite literally, shake the whole system apart unless the
troublesome elements were quickly removed from service. The stability of a system refers to its ability
to remain in synchronism despite disturbances.
During normal operation, the power system is subjected continually to minor disturbances caused bya
multiplicity of load changes (e.g., when consumers switch power on and off). The response of the power
system to these minor disturbances, which is controlled automatically by the individual generator
excitation systems, is a measure of its dynamic stability. However, if a major fault (e.g., the loss of a
bulk power transmission circuit) occurs near a large generating station, the generators on line react
within a fraction of a second, and there are large swings in power flow across the system. The transmis-
sion system interconnecting the generators must be of sufficient capacity to accommodate these
10 The Electric Power System
changes in power flow, Otherwise, a generator may pull out of lock-step, and unless it is quickly taken
out of service the disturbance may spread over the whole interconnected system. The ability of asystem
to remain stable after a major disturbance is known as transient stability. The security of a system
refers to its ability, not only to remain stable after a disturbance, but also to regain an acceptable
operating steady state. The two regional reports of the RCEPP on the need for additional bulk power
facilities in southwestern and eastern Ontario contain a discussion of the security aspects of transmis-
sion planning in Ontario. !
Transmission lines are only one of the numerous elements that comprise a bulk power transmission
system. Important among the others are power transformers, circuit-breakers, buses, synchronous
condensers, static capacitor banks, and shunt reactors. As the name implies, power transformers
change transmission voltages from one level to another. They tend to be huge — a typical modern
transformer to step the voltage down from 500 kV to 230 kV has a mass of about 300 tonnes. Circuit-
breakers are devices that close or open power circuits not only under normal operating conditions but
also in an emergency. They differ from ordinary power switches in that they can interrupt fault
currents that may be at 16 to 20 times the normal operating levels. Buses are, essentially, large copper
or aluminum conductors that are the common connection points for all transmission circuits entering a
transformer station. Synchronous condensers, static capacitor banks, and shunt reactors are installed
at transformer stations to provide reactive power compensation and thereby facilitate voltage and
power factor control across the system.” The cost of the transformers and other transmission equip-
ment at the end of 1977 represented about 9 per cent of Ontario Hydro’s capital investment, compared
with 10 per cent for transmission lines and 61 per cent for generating facilities.
Recently, there has been a revival of interest in high-voltage direct current (HVDC) systems for the
transmission of large amounts of power over great distances. H VDC transmission requires, basically,
two more steps than HVAC transmission. At the generating end, after the AC-generated power has
been transformed to a bulk transmission voltage level, it is converted to DC by rectifiers at the same
voltage, and then transmitted as DC. At the receiving end, the high-voltage DC power is converted
back to AC by inverters, and then the voltage is stepped down, for distribution.
The main advantages of HVDC transmission are:
e Asingle-circuit AC line has three phases and, normally, two skywires (located at the top of the
towers, skywires protect the line against lightning and carry fault currents), whereas a bipolar DC
line has two poles (positive and negative) and, usually, one skywire. The wiring requirements for
HVDC are thus smaller and less expensive.
e Thesmaller number of wires needed for DC overhead lines means that smaller and lighter towers
can be used and that the right of way can be narrower.
e DC transmission permits the interconnection of two power systems that may not otherwise be
capable of parallel operation. For example, the systems of Ontario Hydroand Hydro-Québec, due to
stability considerations, cannot in their present form run in synchronism as one system. To inter-
change power, one party must isolate some of its generating stations from its system, and then
connect those stations to the other system. Due to the same limitation, the Quebec and New Bruns-
wick systems are interconnected by an HVDC link.
e For underwater or underground transmission where cables must be used, the use of AC is quite
limited because, at some critical length, the AC cable becomes loaded with reactive power. Fora 500
kV AC cable, the critical length is about 25 km. This limitation does not exist with DC cables. DC is
particularly useful for underwater transmission over great distances; for example, there is the 32
km 260k V HVDC underwater cable between Vancouver Island and the mainland.
e Thereare nostability problems associated with DC transmission.
HVDC transmission hassome disadvantages also:
e A main disadvantage of DC transmission is the cost of the conversion equipment (rectifiers and
inverters) at the terminals.
e A lack of experience with HVDC andthe complexity of the terminal equipment have resulted in
reliability problems at some HVDC installations. However, with the introduction of solid-state
devices called thyristors, for the rectification and inversion of power, the reliability of HVDC
systems is expected to improve.
° Because of the high initial cost of terminal equipment, overhead H VDC transmission is economi-
cal only for long-distance, point-to-point, bulk power transmission. These distances are in the
order of 650-800 km or more. An example is the Nelson River Project in Manitoba. This project
a
The Electric Power System and Its Components 11
consists of two bipolar HVDC lines, each 890 km long and operating at +450 kV. The first line’s
capacity is 1,620 MW. The second line is operational, but at only half the rated capacity of 1,800
MW; it is expected to be loaded to full capacity by 1981.
Ontario Hydro is studying two applications of HVDC:
e An HVDC interconnection with Hydro-Québec. For the reasons given earlier, an HVAC inter-
connection is not considered practical for this application. The ultimate capacity of this HVDC tie
could be 3,000 MW.
e AnHVDCtransmission line between Ontario Hydro’s East System and its West System. This tie,
with alength of 600 to 800 km, may be economical because it would cross largely unpopulated areas
and thus would not have any load supply along the way. The tie might have a capacity of 1,000 to
2,000 M W in the late 1980s.
The Distribution of Electric Power
Electric power from the bulk power transmission system (at 230 k V and 500 kV) undergoes a number
of voltage transformations before it reaches the ultimate customer. Ontario Hydro classifies them in
three categories: area supply, subtransmission, and distribution. The bulk power at 500 k V or 230 kV is
transformed to 230 kV or 115 kV at the receiving terminal stations, from which it is fed into the area
supply lines (Figure 2.2). These lines, which take power to area-supply transformer stations in or near
cities and towns, are mainly overhead, but in large cities such as Toronto they are often placed under-
ground. At the area-supply transformer stations, power is stepped down from 230 kV or 115 kV to 44
kV, 27.6 kV, or 13.8 kV. These stations are also the delivery points for large industrial customers of
Ontario Hydro, for example, the petrochemical industries in Sarnia. For other customers, power is fed
into the subtransmission lines, which deliver it to distribution stations in or near cities, towns, and
villages.
At the distribution stations, power is stepped down further for supplying to individual customers or
groups of customers. A number of distribution voltages are in use in Ontario. For distribution to groups
of customers, the voltage level is usually 12.48 kV, 8.32 kV, or 4.16 kV. In old residential areas, the
distribution lines at these voltages are usually strung on wood or concrete poles routed along the streets.
In new residential subdivisions they are often placed underground. Distribution transformers are
used along these lines (on poles for overhead lines and on above-ground concrete foundations in metal
casings for underground cables), to step the voltage down to serve individual customers. The voltage
level is 120/240 V for residential, 120/240 V or 120/208 V for small commercial customers, and 600 V
for light industries. The higher of the two voltages available to residential and small commercial
customers is for high-power appliances such as stoves, refrigerators, washers, and dryers. The higher
voltage permits a more efficient utilization of electric energy by such devices.
In Ontario, most of the generation, bulk power transmission, area supply, and subtransmission facili-
ties are owned and operated by Ontario Hydro. Most major urban centres are served by municipal
electricity utilities that buy power from Ontario Hydro and retail it to their customers. However,
Ontario Hydro provides power directly to certain industrial customers,such as pulp and paper, petro-
chemical, and mining companies, and to retail customers in rural areas and communities not served by
municipal electricity utilities.
The End Use of Electricity
F'rom the distribution network of an electric power system, electric energy is supplied to hundreds of
diverse users. In our day-to-day activities, we use electricity for five basic purposes: heating (resistance
heating, induction heating, microwave heating, arc furnaces, and infrared heating), lighting, motors,
electrolytic processes, and electronic devices such as television, radios, amplifiers, and computers. The
nature and amount of each end use varies from one segment of society to another. In the steel industry,
high-power electric motors are used to drive giant rolling mills on a more or less continuous basis. In
contrast, small electric motors are used in the home to power appliances such as blenders, washers, and
dryers, and these are normally used rather infrequently and for short periods.
Traditionally, the end use of electricity, and of energy in general, is classified in four market sectors —
industrial, transport, residential, and commercial (the commercial sector includes all uses that are not
included in the other sectors, e.g., street lighting, office buildings, hospitals, and water works). This
classification reflects the nature and size of the various end uses of electricity. In Ontario, it also reflects
the rate structure to a certain extent.
Se ee a ea ie eT ES st Oy eee eS CS en ee
12. The Electric Power System
In Ontario, and in North America as a whole, almost a quarter of the total energy consumed is used in
the transport sector. However, most of this energy is in the form of diesel oil, kerosene, and gasoline.
The transport sector accounts for an insignificant part of the total electric energy use. In Ontario, for
example, the only major user of electricity in this sector is the public transit system of Toronto. For this
reason, we will include the electric energy consumed by the transport sector with that used in the
commercial sector.
The industrial sector, the largest of the sectors, took more than 40 per cent of the total electric energy
consumed in Ontarioin 1976 (Table 2.4). The largest single end use in the industrial sector is the electric
motor drive, accounting for more than 75 per cent of total industrial consumption. The other industrial
uses of electricity are direct heating, electrolytic processes, and lighting. Manufacturing accounted for
about 80 per cent of the total industrial electric energy consumption, and mining for about 10 percent.
Table 2.4 Ontario Electricity Consumption — 1976
Share of total Share of use
electricity consumption within the sector
End-use sector (%) (%)
Industrial
Direct heat 2.8 6.9
Motor drive 30.6 75.9
Electrolytic processes 2.8 7.0
Lighting 4.] 10.2
Total 40.3 100.0
Commercial
Space heating 2.1 6.8
Water heating 14 45
Air conditioning 8.9 28.8
Lighting and other 18.5 59.9
Total 30.9 100.0
Residential
Space heating 5.8 20.1
Water heating 8.6 29.9
Air conditioning 0.3 1.0
Other appliances 14.1 49.0
Total 28.8 100.0
Ontario total 100.0 =
Source: Royal Commission on Electric Power Planning, “A Race Against Time’’, Interim Report on Nuclear Power in Ontario, September 1978, p. 20.
The commercial sector was the second largest user of electricity in Ontario in 1976, with a 31 per cent
share of the total (Table 2.4). In this sector, electric energy is used almost entirely for heating, cooling,
and lighting. This sector was the fastest-growing one (more than 10 per cent annually) inthe 1960sand
early 1970s. However, in the late 1970s the growth rates declined.
A significant factor contributing to the rapid growth of the commercial sector over the last two decades
has been the trend, in office-block construction, towards the use of large areas of glass, making it
necessary to install more sophisticated and energy thirstier heating and cooling systems than are
required in older buildings. However, engineers and architects have found that in the course of a year,
in a properly designed commercial building, the combination of the heat of the sun falling on the
structure and the heat generated by the occupants and their equipment, including lighting, can exceed
the heat required to maintain a comfortable temperature. Ontario Hydro’s head office in Toronto and
Gulf Oil of Canada’s office in Calgary are examples of such a design. Even during the winter months,
the heat losses and the heat gains in a typical Toronto commercial building are approximately the same.
The share of the residential sector in the consumption of electric energy is roughly 29 per cent (Table
2.4), About one-half of the electricity consumed in this sector is used for space heating and water
heating and the other half is used for lighting and appliances. The growth in the residential use of
electricity in Ontario in the 1960s outstripped the growth in total energy used in this sector. The
principal element in this growth was electric space heating, which grew by 15.5 per cent per year, on the
average, between 1961 and 1971. This rapid growth in electric space heating was a result of the
relatively low price and the convenience of electricity. It was enhanced by Ontario Hydro’s promotion
campaign encouraging everyone to “live better electrically”. The advantages of electric heating are
The Electric Power System and Its Components 18
that it is clean, it requires little maintenance, it responds quickly to the user’s changing needs, and its
installation cost is relatively low. In the 1970s, however, the cost of centrally generated electricity has
risen sharply due to increases in the capital cost of thermal plants and in the prices of coal, oil, natural
gas, and uranium. This, coupled with the low annual load factors of electric space heating, has reduced
the attractiveness of this method of heating in Ontario.
The Nature of the Demand for Electric Energy
For most North American utilities, the demand for electric power (that is, the electricity load) within a
given year varies from hour to hour, from day to day, and from one season to the next. Ontario Hydro’s
load arises from hundreds of diverse end uses — water-heaters, milking machines, household lights,
blenders, chick-brooders, mixers, saws, radios, television sets, furnaces, washers, dryers, refrigerators,
stoves, floodlights, streetcars and subways, boring mills, grinders, rolling mills, electrochemical
processes, etc. Although the patterns of use of the individual devices are very diverse, their combined
use results in total load on the generating system that, while varying over a given period of time, has a
relatively orderly pattern.
Hourly, Daily, and Seasonal Variation
For a working day in December, a working day in July, and a weekend in December, the hourly
variations of load for Ontario Hydro’s total system are as shown in Figure 2.4. For the December
working day, the load increases from a minimum during the early-morning hour’ to a secondary peak
around noon and then to the daily peak in the early-evening hours. The evening peak on a winter’s day
is a result of the living habits and employment patterns of the majority of the population — when people
come home from work at the end of the day they switch on the lights, turn up the heat, and start cooking
supper. For the July working day, the load is more even during the daytime and the daily peak may
occur at almost any time between 10:00 a.m. and 6:00 p.m. The absence of an evening peak in the
summer is explained toa large extent by the reduced need for lighting and heating.
Figure 2.4 also indicates that the daily peak is more than 20 per cent lower in summer than in winter.
This illustrates the seasonal variation of electrical load in Ontario. Ontario Hydro is a winter peaking
utility, and the loads are considerably lower in the summer months than in the winter months. The
seasonal variation in load is more clearly illustrated in Figure 2.5, which shows the daily peak and
minimum loads for the Ontario Hydro East System in 1975. Figures 2.4 and 2.5 also show how the load
can vary within a week. The loads are generally lower at weekends than on weekdays, primarily
because of reduced commercial and industrial activity.
Load Factor
Over a given period, the ratio of the average demand for electric power to the peak demand is called the
“load factor”. The load factor is calculated by dividing the total demand for electric energy (in kilowatt
hours) in a given period by the product of the peak demand (in kilowatts) in the period and the length of
the period (in hours). For Ontario Hydro, typical load factors are:
East System (%) West System (%)
Winter working day 81-87 90-95
Summer working day 84-88 90-96
Calendar year 64-67 77-80
The higher load factors for the West System reflect that area’s greater primary industry loads, such as
paper manufacturing and mining, which tend to operate continuously throughout the week. The sig-
nificantly lower annual load factors (as against daily load factors) reflect the extent of seasonal varia-
tion (Figure 2.5). The electrical load is considerably smaller on a summer day than ona winter’s day. We
will discuss later how the utilities take advantage of this seasonal variation by carrying out the routine
maintenance of their generating equipment during the season of low demand (see Figure 2.5).
Load-Duration Curves
The load shapes shown in Figures 2.4 and 2.5 make it clear that, over a given period, there is a range
within which the load varies. The minimum value is called the base load and the maximum value is
called the peak load. The base load over a period of a year lasts for all 8,760 hours of the year, whereas the
14 The Electric Power System
Fig. 2.4: p. 18
Fig. 2.5: p. 19
Fig. 2.6: p. 20
peak load lasts for less than an hour. Loads that last for less than a year but for more than an hour are
termed intermediate loads.®
for planning purposes, Ontario Hydro finds it useful to utilize the concept of a “load-duration curve’.
The load-duration curve over a given period of time (e.g., a year) is an ordered arrangement of hourly
loads, each consisting of the average load over one clock hour, from the highest to the lowest (Figure
2.6), and it is obtained from the pattern of hourly loads (Figure 2.4) over that period. As Figure 2.6
indicates, Ontario Hydro, for convenience, uses a standardized form of the load-duration curve in
which a given level of load is represented as a percentage of the annual peak, and the duration is
represented as a percentage of the total number of hours in the year. The “20-minute” peak refers to
the 20 minutes during which the total energy demand was the highest, and is the average value of the
load over that peak period. The system should, however, be capable of meeting the momentary peak
that is higher than the 20-minute peak.
A load-duration curve illustrates the relationship between a load and its duration. It may be seen from
Figure 2.6 that in 1977, for 40 per cent of the time (about 3,500 hours), the load in Ontario Hydro’s East
System was greater than 70 per cent of the annual peak. The concept of aload-duration curve is a useful
and important one. The area under an annual load-duration curve is equal to the annual energy de-
mand. From this it also follows that the annual load factor is equal to the average value of the load-
duration curve.
Figure 2.1 Illustration of Load Diversity
System load
System load = Load A + Load B
Ps <P, + Pe
Source: RCEPP.
The Electric Power System and Its Components 15
Figure 2.2 Components of an Electric Power System
Generating stations
Switchyards
ps
=
=)
joe]
im
Bulk power transmission lines
230 kV
Area supply lines
Large users fe
27.6 kV
—_
~ <4
=>
—<=t-
Subtransmission lines
>
ae x
wo NN
—_ fae)
=a co
=
==
co
=e
NS
—
Distribution lines
Distribution transformers
Source: RCEPP.
16 The Electric Power System
120/208 V
>
>
120/240 V
Customers’ loads
BUN}eay J94}0 pul BUIp|ing 40) pajq yeay pue ‘Sasso| Ja}loXa ‘UOIeJQUAB ‘UOI}eipeL jo Ajulew dn apew si 40} payuNodde you %E Ay J9|10q oy] 0} dwind ay} Aq pawsnjas si pue Yasuapuod ay) Z WQIUXF ‘9/61 YEW ‘dd3DY 0} UOIssiwans
J |S ay} SPJ29SID ‘auiquN} a4 Bui}e}0) Aq awos dn saAis 4a]10q a4} Je ABJaus dn syoidy) sales ABlaua auy si yj ‘dwnd ayy Aq 9/9Xo auj punos pue PUNO! PIAOW S! JaJeM PUL WeA}s (4M) PIN|} BUIYJOM ay] :a}0N OUPAH OWE\UQ *,, |Ld!UY9a]—UOIJeIEUaN,, :20uNOS
aye)
pip ipa ge JoLZ 12 JEM 4M Ul Ja}
% Sh eal ,,e4SeM,,
Jo81 40 ainjesadwa} adesane ye aye) 0} papueosip yea}
%00T 19|109 0} jan
Jol? JP Wea}S 4M Ul Jeeq 9}9K) Weals auj
%Z] aaydsowye of
aulqun|
Oe ee re
Jo0PS Je Wea}S 4M Ul Jeo}
Jo}eJaUe5)
% Le Ayouyoal3 JU|q J8MOg paj|anj-|iSso4 @ Ul SMO|4 ABiau3 g°Z aunBIy
17
The Electric Power System and Its Components
Figure 2.4 Ontario Hydro Daily System Load Patterns
18,000
14,000
12,000
10,000
Megawatts
8,000
6,000
4,000
2,000
bam, 3 5 ] 9 ll 1 p.m. 3 5 ] 9 ll l
Time of day
Source: “Electricity Demand’, Ontario Hydro
submission to RCEPP, April 1977, Exhibit 83.
18 The Electric Power System
Jaquiaoaq
Jaqwanoy
4390109 Jaqwia}das
Ay19edeo 3} qgepuadag
Sade]no soueua}UleW pue pauue|d
\sngny
Aine
peo| yead ajnuiw-9z Ajieq
(02 H91UX9 ‘9/67 APW dd30Y 0} UoIssiwigns ONpAY OVEIUQ ‘,,AI|IqGeI|9y,, :30uNDS
\udy * Yue Ayeniga4 Alenuef
-QZ jenuue ay} 40 adejuaoeg
yead aynuiw
peo| wnwiuiw Aj1eq
(G/6[ — WaySAS }Se3 OUPAH OleqUG) peo] Alewiig pue Ajioedey Ajieg g°Z asnsi4
19
mponents
7)
Y
/
The Electric Power System and Its C
Figure 2.6 Annual Percentage Duration Curve of Hourly Demands (Ontario Hydro East System — 1977)
80
10
P
=
F
ee
ae
a.
/
ue
60 4
Percentage of the annual
20-minute peak
20
40
30
20
0 20 40 60 80 100
Percentage of the total number of hours in a year
Source: ‘Bulk Power Facilities — Eastern Ontario”,
Ontario Hydro submission to RCEPP, December 1978,
Exhibit SE-2.
20 The Electric Power System
CHAPTER THREE
Mix of Generating Resources
To be useful, electric energy must, as noted in Chapter 2, be converted from some primary source of
energy, such as coal, uranium, or flowing water. In the context of an electricity generating system,
“mix” refers to the proportions of the various primary energy sources used in the system. Many factors
affect the choice of a generating mix for a given electricity supply system. Some of the important
factors are the pattern of demand for electric energy (see Chapter 2), the comparative costs of various
generation technologies, technical and operating limitations, the availability of primary fuels, the
diversity of the fuel base, and socio-environmental considerations.
Before any discussion of the need for a mix and the factors influencing its selection, the distinction
between capacity mix and energy mix should be explained. It is evident from the load-duration curve in
Figure 2.6 that about 45 per cent of the annual peak load is base load. The system load factor, which is a
measure of the total electric energy requirement, was 65.5 per cent for the Ontario Hydro East System
in 1977. Since the base load lasts for the whole year, the amount of energy associated with the base load
is approximately (45 divided by 65.5 =) 70 per cent of the total energy requirement. Thus, although the
base load component is about 45 per cent of the peak load, the base load energy is 70 per cent of the total
energy requirement. What is left makes up 55 per cent of the peak load but only 30 per cent of the total
energy requirement.
The same is true of the generating plants. A 2,000 MW nuclear plant supplying the base load will have
the same share of the total generating capacity as a 2,000 MW oil-fired plant providing peak load.
However, the amount of energy supplied by the nuclear plant will be a considerably larger fraction of
the total energy supplied than in the case of the oil-fired plant. Thus, it is important to distinguish
between the mix of energy actually being supplied and the mix of generating capacity. Table 3.1
illustrates this distinction by showing the capacity mix and the energy mix of the various electric
energy generation resources that were being used in the Ontario Hydro system in 1978.
Table 3.1 Generating Capacity and Energy Mix: Ontario Hydro System — 19784
Capacity mix Energy mix
Resource type (%) (%)
Hydraulic 28.2 37.4
Nuclear 19.9 30.3
Coal 37.4 28.3
Natural gas 2.6 Re
Oil 11.9 1.8
Total 100.0 100.0
Notes:
a) Does not include purchases.
b) Based on December Dependable Peak Resources.
Sources: Ontario Hydro Power Resources Report — 790201. Ontario Hydro Annual Report 1978.
The hourly, daily, and seasonal variations in the demand for electricity put an important requirement
on the generating system, namely, that it should be able to supply loads lasting from a single moment to
a whole year. In the Ontario Hydro system, due to factors of cost, reliability, and operating characteris-
tics, and because the amount of hydraulic capacity is limited, this requirement can best be met by a
variety, or mix, of generation types, such as coal, oil, gas, and nuclear thermal generation and hydraulic
generation. These and other factors that affect the conventional types of generation are detailed in
Appendix B. This chapter will provide acomparative assessment of these factors.
Comparative Costs of Generation Alternatives
To facilitate discussion, the total cost may be divided among capital cost, operations and maintenance
(O&M) costs, and energy production cost, or fuel cost. Capital cost is defined as the sum of the direct costs
(capital equipment, physical plant, etc.) and the indirect costs (engineering services, construction
camps, ete.) that are needed to design, construct, and commission a project, as well as the interest on
funds spent during construction up to the actual in-service date. It is expressed as dollars per kilowatt
of installed capacity. The O&M costs include the costs of the labour and material required to maintain
The Electric Power System and Its Components 21
and operate a plant, and in the case of nuclear plants also the heavy-water upkeep and any additional
security costs. The O&M costs do not depend to any great extent on the amount of energy produced and
thus are expressed in terms of dollars per kilowatt per year. The fuel cost reflects the cost of ready fuel,
e.g., the cost of fabricated uranium fuel bundles, and is expressed in terms of dollars per kilowatt per
hour of electric energy produced. Table 3.2 provides a comparative cost assessment of the generation
resources at present being used in Ontario. The actual dollar values are outlined in Appendix B, but to
demonstrate the economic suitability of various forms of generation for loads of different durations, a
comparative assessment is useful.
Table 3.2 Comparative Costs of Conventional Generation Technologies Used in Ontario
Generation technology Capital cost Operating cost
Hydroelectric site-dependent very low
Steam thermal
Nuclear high low
Coal medium medium
Oil and natural gas medium high
Combustion turbine low very high
Source: RCEPP.
The capital costs of hydroelectric developments tend to be high, but such plants; once built, are virtually
unaffected by inflation, because the “fuel” is free. The operating costs consist mainly of maintenance
and water-rental charges, which are relatively small. These cost characteristics make hydraulic power
best suited for base-load generation, but the maximum amount of energy that can be generated at any
potential site is limited by the size of the available water supply and the difference in elevation, called
the “head”, through which the water can be made to fall. These factors are determined by the natural
features of the site: the pattern of rainfall and runoff, the topography, and the geology.
The pattern of runoff for some rivers, e.g., the Abitibi and the Mattagami, tends to be highly variable
from season to season. For such rivers, it is generally desirable to develop sufficient water storage
capacity to make it possible to regulate the flow into the turbines so that it corresponds to the variations
in the load (the excellent load-following capability of hydraulic units will be discussed later in this
chapter). On the other hand, for rivers in which the fiow is more or less uniform throughout the year,
e.g., the St. Lawrence, such storage may not be desirable, for navigation reasons. Thus, depending on
the natural pattern of inflow and other considerations, hydraulic plants may be designed for base-load,
intermediate-load, or peak-load operation.
In the case of thermal plants, however, if fuel supplies are assured there are no technical restrictions on
their operation at any desired energy output, except for scheduled or forced outages. Thus, cost becomes
an important consideration in determining their suitability for a load of given duration. Nuclear
plants, being relatively expensive to build and relatively cheap to operate, are being used at present on
as continuous a basis as possible, and thus they supply the base load. Coal-fired generation can be used
for base, intermediate, and peak loads, depending on the cost. Oil- and natural gas-fired plants, having
become very expensive to operate, are used for peaking purposes only, but in emergencies they can be
used at higher capacity factors, thus helping to permit flexibility of operation.
Cost Comparison: Nuclear and Coal
Of the generation technologies that are in prospect up to the end of this century, CANDU-nuclear and
coal-fired steam-thermal generation are the major realistic options for large-scale generation of elec-
tricity in Ontario. The major potential large-scale sites for new hydroelectric development are in
remote areas of northern Ontario, on rivers flowing into James Bay. Any decision to develop these sites
will have to be based on the cost of the individual stations, the costs of bringing the power to southern
Ontario, and the socio-environmental impact. Because different proportions of nuclear and coal-fired
generation are possible in a planned expansion, assessment of the relative economics of these two
options, for loads of varying duration, is relevant. It is possible to summarize such an assessment in a
graphic form using the “‘life cycle” cost data for individual stations. Figure 3.1 shows an assessment
based on the data used in a study carried out for the RCEPP ! (see Appendix B). The comparison is for a
new 4 x 850 MW CANDU station and anew 4 x 750 MW coal-fired station, both coming into service in
1985. The operating life of each station is assumed, for the purpose of calculating discounted cash flow,
to be 80 years from the in-service date. The price of coal for this comparison is based on a mix of eastern
eS a I ce a an lpr gi eee sa he Peal al eae a
22 The Electric Power System
Fig. 3.1: p. 34
U.S. coal and Albertan coal, reaching 50 per cent Albertan by 1995. The fuel cost for nuclear generation
includes the irradiated fuel-management cost, which is estimated by the study to be $19.5/kg of
uranium (in 1986 dollars) for interim storage and $28.5/kg of uranium for geological disposal. The
total ($48/kg of uranium) adds approximately $1/MW-h tothe nuclear fuel cost.
The intercept on the vertical axis in Figure 3.1 is the sum of the capital cost and the discounted 30-year
O&M cost per kilowatt of capacity installed, for each type of plant. The slope of the straight lines is the
discounted 30-year fuel cost in dollars per kilowatt hour. Thus, Figure 3.1 shows that, while the fixed
costs (capital and O&M) of a nuclear station are higher by approximately 75 per cent, its fuel cost is only
about one-fifth that of the coal station. The nuclear station is therefore cheaper in the long run for
annual capacity factors higher than 25 per cent. A sensitivity analysis may be carried out from Figure
3.1 to see what variation in parameters is required for coal and nuclear to be competitive for base-load
operations — say, for an annual capacity factor of 75 per cent. Such an analysis shows that a break-even
annual capacity factor of 75 per cent is achieved if the nuclear capital cost is increased by 100 per cent
(that is, to about three times the coal capital cost), or the uranium price is increased by 300 per cent, or
the coal price is decreased by more than 50 per cent compared with the reference-case assumptions. The
study made for the Commission also showed that, for coal to be competitive with nuclear for base-load
use, the real discount rate would have to be increased from 5.5 per cent per annum in the reference case
to about 17 per cent. It is recognized in the study that the cost of geological disposal of spent nuclear fuel
is quite uncertain, but the estimated cost is such a small part of the total cost that it has to increase by
more than one order of magnitude to offset the advantage of nuclear for base-load generation.
While anything is possible, the variations that are required in a given parameter for a new coal-fired
station to be competitive with a new nuclear station for base-load operation are not likely to occur. The
study therefore appears to be correct in stating that, in terms of the economic costs of new base-load
generation in Ontario, “nuclear generating stations are substantially more attractive than coal-fired
generating stations”. Similar conclusions were arrived at in a January 1979 study by Ontario Hydro’s
System Planning Division, comparing the economic costs of a4 X 750 MW coal-fired station anda4 x
850 MW CANDU nuclear station.” The break-even, lifetime, annual capacity factor calculated in the
study is 35 per cent when a new nuclear station is compared with a new coal station fired with U.S. coal,
and 25 per cent when such a station is compared with a new coal station fired with western Canadian
coal.
While studies comparing the cost of various types of generating stations indicate the economic suitabil-
ity of a station for a given capacity factor, they do not provide an adequate basis for decisions, in a
changing planning environment, about the mix and in-service dates of new stations to be added. The
capacity factor of a plant changes from year to year and may decline over the plant’s life. As new and
more efficient technologies are deployed, the old plants are pushed up in the loading order and may be
operated at considerably lower capacity factors.® This may also happen as a result of an anticipated or
unanticipated decline in load forecast. The in-service dates of the new generating facilities is usually
determined by system reliability considerations and sometimes also by the economics of substituting a
new and more efficient station for the old and expensive one. Also, the cost analysis of single stations
does not take into account the impact of various rates of system expansion on the “front end” of the life
cycle of a plant, e.g., the capital markets, heavy-water plants, uranium- and coal-mining infrastruc-
tures, and coal transportation systems. The economic issues associated with the various generation
technologies in Ontario are discussed in greater detailin Volume 5 of this Report.
The Reliability and Performance of Various Technologies
Of all the types of generating units that are used in Ontario, the hydraulic is the most reliable, with a
forced outage rate of 0.5 per cent and a capability factor of 95.5 per cent.4 Ontario Hydro estimates
show that the reliability of various steam-thermal generation technologies (nuclear, coal, oil, and gas) is
comparable, with capability factors in the 75-80 per cent range. Gas turbine units are the least reliable,
mainly because of their high forced outage rates (about 15 per cent). The overall capability of gas
turbines is estimated to be about 76 per cent.
When an operating unit goes out of service because of a forced outage, it must be replaced by another
unit that will continue to supply the load. The system must have capacity in excess of the peak demand
on any day to meet such contingencies. This capacity is called reserve capacity. When expressed as a
percentage of the peak load, it is referred to as “reserve margin”. The planned maintenance of gener-
ating units in the Ontario Hydro system at present takes advantage of seasonal load variations and
Mix of Generating Resources 23
does not require additional capacity (see Figure 2.5 in Chapter 2). Maintenance is carried out mainly
between March and October when the demand is relatively low. Ontario Hydro’s system requires a
reserve margin of approximately 25 per cent over the annual peak load. However, due to a sharp decline
in the loads from the levels forecast in the early 1970s, the present system has considerable excess
capacity. The installed reserve at the time of the winter peak in January 1979 was 48 per cent, repre-
senting an 18 per cent surplus. Ontario Hydro is trying to sell some of this excess to U.S. utilities. This
issue will be discussed in detail in Chapter 5.
The unreliability of a plant affects the generating mix through the reserve requirements and thus
through the additional cost penalties associated with the plant. The economic cost comparisons given
earlier did not take into account the cost penalties of the relative unreliability of the generating units.
The cost of this unreliability is illustrated by Figure 3.2. It shows the capital cost and the 30-year
discounted O&M cost per kilowatt of the nominal capacity as well as of the load-meeting capability of
nuclear and coal-fired units coming into service in 1985, as a function of unit size. The load-meeting
capability refers to the load a unit could supply with a specified level of reliability, whereas the nominal
capacity is the nameplate rating of the unit. The difference between the cost per kilowatt of load-
meeting capability and the cost per kilowatt of nominal capacity is a measure of the cost of unreliabil-
ity, which is approximately 15 per cent of the cost per kilowatt of nominal capacity for the 500 MW
CANDU and coal-fired units. A higher percentage cost penalty is associated with bigger units due to
their higher forced outage rates and the fact that, for the same size of system, the larger the unit size,
the higher the reserve requirement for a given level of reliability. For the same reason, the load-
meeting capability of a unit is a function both of its forced outage rate and of the size and characteris-
tics of the system with which it is associated.
The information presented in Figure 3.2 is based on studies made by Ontario Hydro in 1975 on the
assumption of a much bigger future system than is being forecast now. To determine the exact effect of
a lower system growth forecast on the load-meeting capability would require a detailed computer
study. Qualitatively, we can say that a smaller system will have a greater impact on the bigger units as
far as a reduction in the load-meeting capability or an increase in the unreliability cost penalty is
concerned.
The effect of the unreliability of plants on the utilization of generating capacity to supply load is
illustrated by Figure 3.3. This figure is based on Ontario Hydro’s generation programme proposed for
1995 in long-range forecast LRF 48 and on Hydro’s 1976 load forecast. Although Figure 3.3 is based on
old information and is representative of only a given future year, its use here is quite appropriate. The
capacity distribution depends on the load-duration curve and on the characteristics of the system’s
generating units (capability factor, size, fuel cost, and maintenance schedule). However, as long as the
mix of stations in the system is more or less unchanged, the shape of capacity distribution will probably
not be significantly different.
The estimated distribution of generating capacity (also known as the plant-duration curve) is related
to the principle of ‘‘merit order loading” to minimize the expected operating costs. It reflects the effects
of forced and scheduled outages and deratings of generating units, which prevent them from operat-
ing without interruptions while supplying the load. Because of these outages and deratings, the units at
the lower end of the loading order (that is, those supplying the base load) are limited to a capacity factor
of less than 100 per cent. To make up for this deficiency in supply to the base load, the units that are
higher in the loading order must generate correspondingly more energy. The areas under the plant-
duration and load-duration curves (see Figure 3.3), therefore, are virtually equal. Also, since the peak
system capacity is more than the peak load (the difference being the reserve capacity), the system
capacity factor, which is a measure of overall capacity utilization, is less than the system load factor. For
example, if the system load factor is 70 per cent and peak capacity is 25 per cent more than peak load,
the system capacity factor is 56 per cent. Note that, in a hypothetical system in which the generating
units are available all the time, the plant-duration and load-duration curves would be identical and the
reserve requirements would be zero.
The availability, and thus the reliability, of Ontario Hydro’s large thermal generating units is an
important factor affecting the reliability of any planned generation programme. The Sierra Club of
Ontario, in its submission to the Commission in September 1978, expressed concern about the decreas-
ing availability of Ontario Hydro’s thermal units and about its forecast of availability of future large
units:
Clearly there has been an alarming decline in the availability of thermal generating units. The
eee ees ae re eae eee eo a ee
24 The Electric Power System
Fig. 3.2: p. 34
Fig. 3.3: p. 35
Fig. 3.4: p. 36
answer is not more spare capacity but identification of the causes of this alarming trend and return to
1970 levels of availability. The Club suspects that the rapid scale-up of plants has been a contributing
factor, 1.e., larger units, immature designs, and failures in quality control.
With 24 per cent of its thermal units unavailable in 1975, including designs widely used and proven,
[Ontario] Hydro’s projection of 17 per cent unavailable (12 per cent forced plus 5 per cent mainte-
nance) on mammoth 1,250 MW units must be viewed with great skepticism.”
‘Alarming decline in the availability” refers to the second column of Table 3.3, which shows the aver-
age incapability of thermal units as a percentage of the primary peak demand. The Sierra Club’s
interpretation of these numbers appears to be misleading. An increase in the average incapability, as a
percentage of peak demand, would be expected because of a sharp penetration of the system by thermal
units, and is thus a result of the changing system mix. As the table shows, thermal capacity increased
from 4,985 MW in 1970 to 10,577 MW in 1975. A relevant measure of the unavailability of the thermal
- capacity is given by its incapability, expressed in terms of the total thermal capacity and not in terms of
peak demand. This is shown in the last column of Table 3.8. It should be pointed out that many of
Ontario Hydro’s 500 MW-unit thermal stations were put into service between 1970 and 1975 (Picker-
ing 1-4, Lambton 2-4, and Nanticoke 1-5). Thus, the unavailability of the thermal capacity during these
years is representative of the incapability of these units during their early years of service, which is
forecast by Ontario Hydro to be in the order of 30 per cent and not 17 per cent as mentioned by the
Sierra Club. Ontario Hydro’s incapability projections include forced and maintenance outage rates as
well asa planned outage rate, which is about 10 per cent (see Chapter 4).
Table 3.3 Total Incapability of Ontario Hydro’s Steam-Thermal Capacity (1970-75)
Year Average incapability Primary peak Average Average thermal Average incapability
as a percentage of demand incapability dependable capacity as a percentage
primary peak demand? (MW) (MW) (MW) of thermal capacity
1970 11 11,289 1,242 4,985 25
1971 16 11,534 1,845 6,395 29
1972 17 12,739 2,165 7,503 29
1973 18 13,606 2,400 8,742 27
1974 Ze 13,538 2,978 10,171 29
1975 24 14,513 3,483 10,577 33
Note a) Although these data are obtained from page 21 of Ontario Hydro’s submission to the RCEPP entitled ‘‘Reliability’’, the incapability for 1973 is changed
from 21 per cent to 18 per cent. The average incapability corresponding to 21 per cent of primary peak is 2,860 MW, which appears to be very high when
compared with Chart 14D of Appendix 10-E of this submission, which shows the variation over time of the total incapability of Ontario Hydro’s thermal capacity
in 1973. From this plot, the average incapability appears to be in the order of 2,400 MW, which is approximately 18 per cent of the primary peak demand.
The question of availability is even more significant for the CANDU nuclear units, which are highly
capital-intensive and suitable for base-load operation because of their low fuelling costs. Because of the
relatively few CANDU reactor years of operation, it is difficult to make a statistically meaningful
comparison between their actual availability and their forecast availability. Our analysis is based on
the experience (approximately 30 reactor years) with the 4 x 500 MW Pickering A station, whose first
unit was in service in 1971. Table 3.4 shows the annual capacity factors (ACFs) of the four units in
various years of service. The yearly averages are plotted in Figure 3.4. Also shown in Figure 3.4 is the
1975 Ontario Hydro forecast of capability for use in system planning studies (see Table B.1 in Appen-
dix B). It may be seen that the capability in the first four years of service has been less than expected,
whereas in the later years it is better than expected. In terms of the lifetime average capability, both
Unit land Unit 2 have done better than Ontario Hydro’s forecast average of 77.7 per cent for an eight-
year-old unit and 77.4 per cent for a seven-year-old unit. Unit 3’s record is not near the expectation,
whereas Unit 4’s capability is marginally below the forecast average of 76.9 per cent for a six-year-old
unit.
On the basis of limited experience with the Pickering A station, it may be concluded that the CANDU
reactors should be able to maintain base-load capacity factors. This conclusion is partly based on the
clear positive trend in the capability factors with the years of operating experience depicted in Figure
3.4. As far as the availability of the larger CANDU units (for example, the 850 MW Bruce and Darling-
ton units) is concerned, no meaningful analysis is possible because of insufficient operating experi-
ence.° It should be pointed out that Ontario Hydro’s forecast predicts a lower capability for the larger
units (77.4 per cent for the 850 MW units, compared with 80.1 per cent for the 500 MW units). An
analysis of the operating experience with the light-water reactors (LWRs) in the U.S. indicates “a
Mix of Generating Resources 25
Table 3.4 Pickering A Generating Station — Annual Capacity Factors
Year of service 1 2 3 4 5 6 7 8 Lifetime unit average
Unit 1 80 45 92.5 72 80 92.8 85.6 95.1 80.4
Unit 2 50 69 88.4 86 93.2 91.0 84.3 - 80.3
Unit 3 30 85.1 42.7 575 93.9 95.6 82.2 - 69.6
Unit 4 90.1 93.9 23.8 68.4 90.8 89.6 - - 76.1
Yearly average 59.4 73,3 61.9 71.0 89.5 92.3 84.0 95.1 -
Notes:
1) Unit capacity factors and the yearly average for the first year of service are based on operation for only part of the year. .
2) The significantly low capacity factors of unit 1 in its second year of service and of units 2 and 3 in their first year of service are in part due to a 1972 strike. This
is not attributable to unit design and thus to the unit’s capability.
Source: Ontario Hydro, ‘Reliability’, submission to RCEPP, May 1976, Exhibit 20; and other Ontario Hydro information.
distinctive trend in the availability as a function of plant [unit]size”.’ The U.S. data for the pressurized
light-water reactor (PWR) category indicate average availabilities of 79 per cent and 72 per cent for
the 500 MW and 850 MW units, respectively.
The choice of the optimum unit size on economic grounds is quite sensitive to the rate at which the
availability changes with the size. At present, no universally acceptable measures of this rate are
available.
In its Interim Report on Nuclear Power in Ontario, the RCEPP concluded that “the 1,250 MW CANDU
reactors should not be part of Ontario Hydro’s system expansion programme before the turn of the
century”. This conclusion is further underlined by the declining load forecasts by Ontario Hydro, and
by the uncertainty about the capability of these units. Concomitantly, it is noted that Ontario Hydro
decided in 1979 to stop all work on the 1,250 MW unit.®
The Operating Characteristics of Various Technologies
Operating characteristics are an important technical consideration in any assessment of generating
mix. The operating characteristics of a generating unit refer primarily to its shut-down and start-up
characteristics, its ability to operate at reduced loadings, and its load-following capability, that is, its
ability to change its output continuously in response to the changing load.
Hydroelectric units have the best operating characteristics. Fach unit ean be adapted for any duration
of load. Cost considerations and natural features generally determine the type of load for which a
hydraulic site is developed.
CANDU nuclear units of current design, because of their high capital and low operating costs, are best
suited for relatively continuous operation, that is, for supplying base loads. Ontario Hydro’s evidence to
the RCEPP indicated that these units can be operated at lower capacity factors by reducing their
output overnight by up to 50 per cent and by weekend shut-downs. However, CANDU units have not
been especially designed for load-following. The nuclear plants currently committed by Ontario Hydro
(up to and including the Darlington Generating Station) are expected to operate only at base-load
capacity factors until the end of the century. This situation may change if the load forecasts drop
further and the committed nuclear programme is not postponed.
Currently, fossil-steam units are used in Ontario Hydro’s system to supply base load as well as interme-
diate and peak loads, and reserve. The fossil-steam units being considered by Ontario Hydro in its long-
range generation plans are best suited, from an operational viewpoint, for the base and intermediate
loads, but can also provide peak loads and reserve.
Although gas-turbine units have fast start-up and shut-down characteristics, f requent starts and shut-
downs tend to increase their maintenance cost. For the Ontario Hydro system, on the basis of current
prices, they are economical for peaking and reserve duty, although they could be operated at higher
capacity factors if the need arose and fuel was available.
The Spectrum of Conventional Technologies
It is possible to combine the factors of cost, reliability, and operating characteristics to determine the
efficient operating range of each of the electric power technologies. Figure 3.5 illustrates this range ina
general way for the conventional generation alternatives in Ontario, including purchased power and
storage. As the figure shows, all generation technologies are limited to annual capacity factors of less
26 The Electric Power System
Fig. 3.5: p. 36
than 100 per cent due to the reliability constraints (forced and scheduled outages). However, this may
not apply to the power purchased under a contract from another system. The seller of such power may be
bound by the contract to deliver power at a 100 per cent capacity factor. The seller can satisfy this
requirement by providing sufficient reserve capacity in his system.
Fuel Requirements and Supply
Ontario Hydro’s current contract with Petrosar Ltd. to supply 7.3 million barrels of low-sulphur resid-
ual oil per year to 1992 is adequate for the 1,116 MW Lennox Generating Station operating at an
annual capacity factor of 50 per cent. (The actual installed capacity of the four-unit Lennox station is
2,232 MW, but two units were taken out of service early in 1979 due to less-than-expected loads.) In its
long-range forecast LRF 48A, Hydro indicated that it planned to build one more oil-fired station, at
Wesleyville, with a capacity of 2,164 MW, but the planned size was cut in half in view of the lower 1978
load forecast. In February 1979, with a further drop in expected load growth, Hydro announced that it
would “mothball” the Wesleyville Generating Station, to reduce excess capacity. All components for
that station will be delivered and stored on site until the early 1990s. Of an estimated total cost of $660
million, about $380 million has been committed. Hydro is also planning to consider the advantages of
converting the two Wesleyville units to coal, or a combination of coal and oil, before the station is
eventually completed. As a result of the cutbacks, Hydro hopes to reduce by one-half the residual oil
contract amount of 7.3 million barrels a year over the period ending in 1992. Hydro is not planning any
other oil-fired steam-thermal stations.
The future prospects of natural gas supplies for the Hearn Generating Station are good. Its four gas-
fired 100 MW units have been taken out of service, in addition to the two Lennox units, due to reduced
expectations of future loads. These six units could be brought back into service on short notice. The
remaining four 200 MW units at Hearn are fired with natural gas in summer only. Hearn’s require-
ments are estimated at 10 billion cubic feet per year to 1988 and 5 billion cubic feet per year thereafter.
These are adequate for Hearn in its role as a peaking or reserve station. In view of the relatively small
requirements for oil and gas for the Ontario Hydro facilities, their supply in the future is not a major
concern.
The oil- and gas-fired units in Ontario are economical at current prices only for peaking and reserve.
Their future use for these purposes may be considerably reduced by the following factors. Ontario
Hydro’s load forecasts have declined from an average annual growth rate to the year 2000 of about 7
per cent in 1975 to one of about 4.5 per cent in 1979. If the load were to grow at 4.5 per cent, the existing
capacity of peaking- and intermediate-load hydraulic generation of about 3,000 MW would provide a
significantly larger share of the peaking needs. Moreover, Hydro plans to develop as much as 2,000 MW
of new hydraulic capacity at 17 sites, with a total average annual output of 523 MW, equivalent to an
annual capacity factor of 26 per cent. This new capacity will supplement Ontario Hydro’s peaking
resources. The total cost of the development is estimated at $1.4 billion (in 1977 dollars). Ontario Hydro
also announced in July 1978 its intention to pursue a load-management programme aimed primarily at
flattening the peak on the thermal generating resources, that is, the part of the peak load that is not
supplied by hydraulic generation. The load-management target for 1985 is 500 MW, rising to 1,300
MW in 1992 and to about 2,000 MW by the end of the century.
In Ontario Hydro’s system, the coal-fired stations constitute about 37 per cent of the total capacity, and
they generated 28 per cent of the total electric energy in 1978. They make up asignificant proportion of
the base-load capacity, at present. Because of the economic advantage of nuclear stations for base load,
all generating facilities being built for the 1980s are nuclear (Pickering B, Bruce B, and Darlington),
except for a 300 MW extension to the Thunder Bay station and the 400 MW Atikokan station in the
West System. With the increasing role of nuclear in supplying base-load energy, the demand for coal is
projected to grow modestly from the 1978 level of 9 million tonnes to about 12 million tonnes per year by
the end of the century, on the basis of Hydro’s 1979 load forecast. Until the late 1980s, Hydro’s problem
will be not securing supplies of coal but rather reducing the considerable oversupply (see Figure 7.4 in
Chapter 7).
Until recently, Ontario Hydro purchased all of its coal from markets in the eastern U.S. Hydro believes
that the reliability of supply from existing U.S. sources is good, but that various factors, including the
ability of U.S. mines to satisfy both the increasing domestic requirements and exports, could affect the
Mix of Generating Resources 27
future supply to Ontario. This factor may have particular significance in the light of a U.S. administra-
tion proposal requiring the U.S. electricity utilities to cut their oil consumption in half by 1990 by
switching their boilers to coal.
To lessen somewhat its dependence on U.S. coal, Ontario Hydro has contracted for about 2.5 million
tonnes of bituminous coal per year from Alberta and British Columbia and about 0.9 million tonnes of
lignite per year from Saskatchewan. The transportation system for this supply consists of unit-train
movement to Thunder Bay and a terminal at Thunder Bay to transfer coal from trains to lake vessels
for shipment to the Nanticoke Generating Station in southern Ontario. Lignite will be unloaded at
Thunder Bay and transferred by conveyor belts to the generating station on nearby Mission Island.
The bituminous coal will be blended at Nanticoke with U.S. coal (a 50-50 mixture) prior to use there.
Movement of the bituminous coal by the integrated transportation system began in 1978 and about 0.5
million tonnes were delivered to Thunder Bay by the end of 1978. Deliveries are expected to reach the
full contracted amount in 1980. First shipments of the lignite are also expected in 1980. The contracts
for western Canadian coal expire in 1993.
Preliminary estimates of the capital investment required for the new system are in excess of $422
million. Ontario Hydro’s share is approximately $79 million, to cover manufacturing of the railroad
equipment and construction of the Nanticoke blending terminal. Other costs, being contributed by the
companies involved, are $133 million in mine development, $60 million for the Thunder Bay terminal,
$90 million towards the improvement of railroad facilities, and approximately $60 million for
shipbuilding.
Each unit-train trip will carry about 9,000 tonnes of coal (100 gondola cars with a capacity of 90 tonnes
of coal each). Thus, about five round trips per week will be necessary for the bituminous coal and two
round trips per week for the lignite. In order to provide upward flexibility, the design of the Thunder
Bay terminal is such that its initial capacity of 2.7 million tonnes of coal per year could be doubled to 5.4
million tonnes. Transport capacity could be increased by deploying more unit-trains and lake carriers.
Western Canadian coal has both advantages and disadvantages over coal from the eastern U.S. The
disadvantages lie primarily in its quality. Bituminous thermal coals in western Canada have lower heat
content (11,000 BTU /Ib., compared with 13,000 BTU /Ib. for U.S. coal), considerably higher ash content,
and sometimes a lower volatile content, which affects combustion stability.’ Saskatchewan lignite, with
only 7,000 BTU /Ib., has a very high moisture content — as muchas 34 per cent of total weight, compared
with 6 per cent in bituminous coal. These factors and the increased transportation and handling costs
make western Canadian coal 40-50 per cent more expensive than current supplies of U.S. coal. From the
viewpoint of quality, the advantage of western Canadian coal lies in its low sulphur content (0.5 per
cent compared with 2.5 per cent for the U.S. bituminous) and this is an asset in terms of air quality and
environmental pollution. An increased reliance on Canadian sources will increase the security of sup-
ply to Ontario Hydro and have a positive impact on Canada’s balance of payments. The investment ina
Canadian venture will also provide a direct stimulus to the Canadian economy by creating jobs, and this
will mean cash flows into the provincial economies.
Ontario Hydro’s existing CANDU nuclear capacity is 5,248 MW. Another 8,612 MW (Pickering B,
Bruce B, and Darlington) is under construction or committed and is scheduled to be in service by 1990,
increasing the total nuclear capacity to 13,860 MW. At an annual capacity of 80 per cent, the 30-year
uranium requirement of this capacity is about 58,000 tonnes. In February 1978, Ontario Hydro re-
ceived the provincial government’s approval to enter into uranium supply contracts with Denison
Mines Limited and Preston Mines Limited. These contracts call for the delivery of 76,200 tonnes of
uranium beginning in 1980 and continuing through the year 2020 — 48,500 tonnes from Denison over
the period 1980-2011 and 27,700 tonnes from Preston over the period 1984-2020. Ontario Hydro has
other contracts for about 5,000 tonnes of uranium. The excess contracted supply of about 23,000 tonnes
is adequate for the 30-year requirement of 5,400 MW of nuclear capacity beyond the currently commit-
ted programme. Table 3.5 gives the latest estimates by Energy, Mines and Resources Canada of the
uranium reserves in Canada. Ontario’s share of the total of the measured, indicated, and inferred
categories is about 68 per cent, that is, about 365,000 tonnes, and of the prognosticated category it is 42
per cent, that is, about 180,000 tonnes. Although Ontario’s share for each of the measured, indicated,
and inferred categories is not available for the latest estimates, it was 60 per cent, 70 per cent, and 74
per cent, respectively, in the 1977 assessment of uranium reserves. The annual supply-and-demand
projections of uranium for Ontario Hydro’s expansion programme, corresponding to its 1979 load
28 The Electric Power System
Fig. 3.6: p. 37
forecast as well as to a lower load forecast of 3 per cent average annual growth to the year 2000, are
discussed in Chapter 7.
Table 3.5 1978 Estimates of Canada’s Uranium Resources (thousands of tonnes of uranium)
Reasonably assured Estimated additional
Measured Indicated Inferred Prognosticated
Up to $125/kgU 76 139 223 147
$125 to $175/kgU 4 16 79 279
Total 80 155 302 426
Source: “1978 Assessment of Canada’s Uranium Supply and Demand”’, Energy, Mines and Resources Canada. Report EP79-3, June 1979.
Heavy-Water Supply and Demand
Ontario Hydro has three heavy-water plants, operating or under construction - BHWP-A, BHWP-B,
and BHWP-D — allat the Bruce nuclear complex (see Appendix B). BHWP-A is operating and BH WP-
B is expected to be commissioned by 1980. Only half of BH WP-D is being constructed, and no decision
has yet been made on commissioning.
The supply-and-demand projections for heavy water are shown in Figure 3.6. The supply projections
are based on the operation of BHWP-A and BHWP-B only. Both supply and demand are cumulative,
and not annual, quantities. The demand is made up of the central inventory for new reactors, and
make-up for losses during operation. As may be seen from Figure 3.6, a surplus in heavy water will
develop during the 1980s under Ontario Hydro’s committed programme. This surplus will be about
4,000 tonnes of heavy water by 1990, assuming the dependable supply, and 6,000 tonnes, assuming the
probable supply. If the expansion of the nuclear capacity in the 1990s continues according to Hydro’s
1979 plan (11,450 MW of additional nuclear capacity), the demand will keep up with the dependable
supply. The surplus with respect to the dependable supply will stay at 4,000 tonnes, but with respect to
probable supply it will increase to 8,000 tonnes by 2000. A tonne of heavy water provides sufficient
central inventory for 1MW of new CANDU capacity. Another way to interpret the surplus is to say that
the heavy-water supply capability would permit advancing the whole nuclear programme by up to 3.5
years if required. The figure also shows the heavy-water demand if no expansion of nuclear capacity
beyond Darlington takes place, for example, under a 8 per cent load-growth scenario. The surplus in the
year 2000, under these assumptions, would be about 15,000 tonnes on the basis of dependable supply.
The above analysis shows that the committed heavy-water production capability is more than sufficient
to supply the requirements to the year 2000. It may be concluded that a provision at this time, or in the
near future, for the expansion of the heavy-water capability is not needed. Indeed, with BHWP-D, the
supply surpluses will increase.
Lead Time
The time required to bring a new generating facility into commercial service, called the lead time, has
become a matter of concern to modern power-system planners. In Ontario, the total lead time must
allow for four major steps. The first step is the identification of a site for the new station and completion
of the necessary procedures, including public participation, for the purchase of the site. The second step
is a detailed geological investigation, an environmental and community impact assessment, and the
necessary preliminary engineering work so that approval for the project may be sought. If a project is
approved, the third step, the preparation of the site, istaken; this includes grading, topsoil removal, and
subsoil preparation. This is followed by the last step, which is the detailed design and on-site construc-
tion of the facility. The typical durations of each of these four steps for various types of thermal electric
generating stations in Ontario are shown in Table 3.6. The lead time for hydraulic stations varies,
depending on the site, from eight to 15 years,
The issue of lead time has become extremely significant because of the increasing uncertainty associ-
ated with the forecasting of loads. Lead-time constraints may, themselves, eliminate from considera-
tion facilities that cannot be put into service by the time they will be required. To make the planning
process responsive to the changing predictions of future loads, it is necessary to reduce lead times as
much as possible. Ontario Hydro informed the Commission during the public information hearings in
1976 that it was investigating methods of reducing lead times. Details of these investigations were
sought by the Commission and were to have been provided by Hydro in a separate submission.!° This
submission was not received.
Mix of Generating Resources 29
Table 3.6 Representative Lead Times (in years) of New Thermal-Electric Generating Stations
Fossil-steam
Nuclear 750 MW 200 MW Combustion
850 MW units units units turbine units
(years) (years) (years) (years)
Investigations and public participation culminating
in approval to acquire a specific site 2-3 2-3 2-3 1
Specific site investigation and public participation,
and preliminary engineering culminating in project
release 3 3 3 ]
Site preparation 1-3 1-3 1-3 -
Detailed design and on-site construction, up to
in-service date of the first generating unit 5,5 45 3.5 l
Total lead time 11.5-14.5 10.5-13.5 9.5-12.5 3
Source: Ontario Hydro, “Generation Planning Processes’’, submission to RCEPP, May 1976, Exhibit 2].
However, Ontario Hydro testified during the public information hearings that a reduction in lead time
in the third and fourth steps — site preparation, and design and construction — is not possible. Im-
provements in lead time are possible, according to Hydro, through the “project-on-the-shelf” concept
by which sites are “banked” in advance by the completion of step one, that is, site acquisition, and
certain activities in step two are carried out — up to but not including project approval, that is, site
investigation and preliminary engineering work. This concept of site-banking could reduce the lead
time from 11-14 years to8-10 years for a nuclear station, from 10-13 years to 7-9 years fora large fossil-
fuelled station, and from 9-12 years to 6-8 years for a smaller fossil-fuelled station. The cost of complet-
ing step one and part of step two could be $10-20 million — it was about $15 million for the Darlington
station. While these costs are not insignificant, they are less than 1 per cent of the total capital cost of
the project.
The Honourable Darcy McKeough, then Minister of Treasury, Economics, and Intergovernmental
Affairs, in his appearance before the Commission during the public information hearings in 1976,
outlined his Ministry’s policy with respect to reducing lead times, and endorsed Ontario Hydro’s posi-
tion on the project-on-the-shelf concept and site-banking:
It seems to me that if we have to look at everything in the context of a lead time of 12 years we are
putting ourselves into a position we don’t need to be in. We should, if I can put it simply, have a
number of projects on the shelf in which the first six years is out of the way.... There would be a
problem here to some extent of credibility both for Hydro and the government of raising some
expectations or fears that might prove to be unnecessary, but it seems to me [that]if we can get twoor
three sites ahead and on the shelf, then we [can] shorten that whole period so that we can go fullouton
the last six yearsif that isrequired.!!
The Sierra Club of Ontario, in its submission to the Commission in September 1978, supported the
concept of site-banking and recommended that “the Commission give favourable consideration to
land-banking as a significant aid to sound electric power planning”.!* While noting that “land-bank-
ing, properly administered, appears to be a reasonable ‘insurance policy”’, the club also proposed a set of
guidelines for land-banking. The guidelines encourage public participation, environmental assess-
ment, the use of multi-purpose transmission corridors, and the use of a generating site for a range of
diverse options, but they caution against activities that may not be in the public interest:
Land-banking should be undertaken to improve planning choices, facilitate environmental assess-
ment and shorten the lead time between final identification of ‘tneed” for generation and plant
commissioning. It should not be undertaken asa land speculation activity.}°
The Sierra Club also recommended other means to reduce lead times, including the use of proven
designs, the use of smaller-scale projects that create less conflict, and the establishment of a clear
framework for decisions and approvals in order to minimize uncertainty.
It must be remembered that site-banking reduces the effective lead time only from the viewpoint of an
upward flexibility in system planning, thatis, only when the demand happens to be more than forecast.
Ontario Hydro is expected to have considerable excess generating capacity until at least the late 1980s,
on the basis of its 1979 load forecast. Volume 8 of this Report, which discusses the factors affecting the
30 The Electric Power System
demand for electricity, indicates that Ontario Hydro’s 1979 load forecast may be too high. Further-
more, there is a potential to expand the generating capacity, especially nuclear, at some of Ontario
Hydro’s existing sites, e.g., Lennox, Darlington, and Wesleyville.!4 On the basis of Ontario Hydro’s
estimates of the “probable maximum capacity”, these three sites could accommodate an additional
thermal capacity (beyond the currently committed programme) in the order of 18,000 MW, which is
more than Ontario Hydro’s planned thermal capacity of 14,200 MW for the 1990s. Hydro has pointed
out that the probable maximum capacity could be influenced by economic, socio-environmental, and
technical factors.
The precedin g discussion indicates that there is sufficient time, at present, for public participation and
environmental assessment with regard to any new site proposal by Ontario Hydro. In the longer term
however, site-banking may play an important role in enhancing the flexibility of planning.
’
System Considerations
It is clear that many of the key variables that affect the planning of asystem, such asthe future demand
for electricity, the availability of resources, and project in-service dates, are characterized by consider-
able uncertainty. The behaviour of the system in an uncertain future, in terms of its ability to survive
disturbances, depends on its resiliency. A resilient system should be able to survive the disturbances
and adapt to the changed conditions. The ideal way to cope with an uncertainty is to recognize it,
estimate the system’s probable behaviour in response to it, and allow for it in the design of the system.
An example of a procedure for coping with uncertainty is the generating reserve margin, which
incorporates an allowance for component failures. Various reliability models study the effect of such
failures on the reliability of the electricity supply, and, on the basis of a specified reliability criterion,
estimate the amount of reserve capacity required.
The resiliency of a system is enhanced if the system is made up of diverse components. It is desirable to
incorporate diversity in the design of a system so that a breakdown of one major component will not
necessarily cause the failure of the whole system. The same concept is applicable when a mix of genera-
tion technologies and primary fuels is chosen for an electric power system. If, for instance, a system
were to be based exclusively on one source of primary energy, its whole operation could be crippled by a
single event, such as a major drought, a prolonged strike of coal-miners, or a catastrophic accident in a
nuclear plant.
An exception to this rule is seen in some systems that are largely hydraulic (such as Hydro-Québec).
They usually have large storage capacity in lakes and a relatively small variation in annual precipita-
tion, so that the chance of a major shortfall in water supply is very low. There is, of course, the possibility
of a major outage caused by a design or operations error during the construction of the dams, or in the
turbo-generators. However, there are usually many generating units per dam in a system, and the
dams are distributed over several rivers, so that many fluctuations average out. Moreover, these are
mature and well-established technologies, and the possibility of a major problem is much less than in a
system that is dependent on emerging (or immature) technologies.
Although the Ontario Hydro system is well balanced at present among coal, nuclear, and hydraulic
capacity, it is important to note the implications of the increasingly complementary nature of the
largely hydraulic generating system in Quebec and the increasingly thermal generating system in
Ontario. For example, a major fault in boiler design or manufacturing might require more than the
usual number of Ontario Hydro’s thermal units to be out of service for maintenance simultaneously.
This sort of problem does not much affect Hydro-Québec, and so a mutual assistance agreement with
Hydro-Québee would give Ontario a more resilient system. Similarly, the rivers that supply the bulk of
Hydro-Québec’s and Ontario Hydro’s power come from different watersheds. Although the Quebec
watersheds are replenished by the same east- or northeast-moving weather pattern that replenishes
Ontario’s watersheds, most of them are north of those in Ontario. Low-water conditions are unlikely to
occur simultaneously in both provinces, so that in most circumstances the thermal and hydraulic gener-
ation in Ontario would be available to assist Quebec.
It must be pointed out that diversity of the fuel base is only one factor in the choice of a mix. The
qualitative nature of this factor makes it difficult to incorporate it in quantitative decision-making
models. However, a primary objective of generating mix studies is long-term reliability of the supply of
energy, and soit is prudent to consider the effect on the system of interruptions tothe energy supply.
Mix of Generating Resources 81
Ontario Hydro’s Practice in Assessing Generating Mix
The methods of quantitative analysis that are used by Ontario Hydro to assess generating mix incorpo-
rate factors such as economic costs, reliability, operating limitations, and capital or fuel-supply con-
straints. Ontario Hydro uses its judgement in the weighing of other considerations, such as socio-
economic factors, fuel flexibility and security, and technical obsolescence.
As mentioned earlier, an economic cost comparison of the CANDU-nuclear and coal-fired power plants
indicates a break-even annual capacity factor of about 40 per cent based on U.S. bituminous coal, and
one of about 30 per cent based on western Canadian bituminous coal. This implies that the CANDU
units are cost-effective when supplying loads lasting for as little as 30-40 per cent of the year. If cost
economics were the sole consideration in the determination of the mix, the long-term share of nuclear
and existing base-load supply could be as high as 70 per cent of system capacity and 90 per cent of
system energy. This follows from the fact that about 90 per cent of the annual energy demand is
contributed by loads prevailing more than 40 per cent of the time (see Figure 2.6 in Chapter 2).
As will be shown in Chapter 7, an analysis of Ontario Hydro’s previous and current generation expan-
sion programmes indicates that the planned long-term share of the nuclear and base-load hydraulic
stations is roughly 60 per cent of the system capacity and 75 per cent of the energy. The difference
between the planned mix and the mix dictated purely by cost economics is a measure of the weighing by
Ontario Hydro planners of the benefits of diversity in the generating mix and the operating limita-
tions of the CANDU reactors at low capacity factors:!°
... there is a concern about fuel supply diversity and flexibility; a very real question: Should we go [all
nuclear] if dollar and cents economics show we should be all nuclear? The dollar and cents [economics
of] nuclear doesn’t take into account flexibility and fuel security.
Our system demand capacity factor couldn’t tolerate ...that amount of nuclear... just couldn’t do it.
The weekly and weekend and seasonal variations in load are such that we would have to examine how
much load cycling the nuclear units have got to do.... Whether or not they can meet the load increase
in two hours every morning, might be another constraint. !®
Ontario Hydro believes that the ability of the generating system to meet unpredictable conditions is
improved if a thermal generating unit is able to use various alternate fuels. A good example of such
flexibility is the R.L. Hearn Generating Station in Toronto. Built in the early 1950s, it was designed to
burn coal, but in 1971, under new regulatory standards on environmental pollution, Ontario Hydro
carried out alterations enabling it to burn either coal or natural gas. At the same time, Hydro also
believes that it is uneconomic to design a plant initially to burn arange of fuels.
Most fossil-steam units can burn only a small range of alternative fuels. This is because their capital
cost must be increased substantially if they are designed to use a wide range of fuels; cost analysis
indicates that it may be less costly to rebuild a boiler to meet a future major change in fuel than to
spend larger initial sums of money to design the boiler .. . for use of a wide range of fuels.!”
However, Ontario Hydro acknowledged before the Commission in November 1978 that, while it is
relatively simple to convert an existing coal-fired station to use oil or natural gas (as was the case with
Hearn), it may not be technically feasible to do the reverse. This is because of the design of coal-fired
boilers and associated coal-handling equipment such as pulverizers and conveyors. Burning coal pro-
duces much ash, the handling of which requires a different type of boiler bottom. Furthermore, there
may be limited space for coal storage at an existing oil- or gas-fired station. An estimate of the increase
in capital cost associated with building a boiler to burn oil as well as coal is 10-15 per cent of the cost of
the oil-fired plant, according toa study done by Ontario Hydroin 1973.18
At present, oil and natural gas play a minor role in Ontario Hydro’s system (see Table 3.1), and Ontario
Hydro plans to build only one more oil-fired station (the 1,082 MW Wesleyville Station). Since oil-fired
stations cannot be converted to burn coal and it seems quite unlikely that coal will be replaced by oil or
natural gas in Ontario, the ability of fossil-steam units to burn other fossil fuels might not enhance the
flexibility of Ontario Hydro’s system to any significant extent. What may be more important is the
ability of fossil-fuelled stations to burn biomass-derived fuels.
Operating flexibility is desirable, in the face of uncertain future load patterns. Due to the long lead
times and relative operating inflexibility of large fossil-steam and CANDU units, operating flexibility
of the system as a whole can be enhanced by the incorporation of short-lead-time storage plants:
Both large fossil-steam and CANDU-PHW units have potential problems if they are operated on a
highly cycled loading pattern. If changing conditions of future loads and generation tend to increase
382 The Electric Power System
the cycling and to decrease annual capacity factors, the solution might be the introduction of energy
storage schemes which would result in meeting cycling requirements and at the same time would
preserve high annual capacity factors on the large thermal units.”
Any consideration of storage schemes must be based on factors such as their lead times, capital cost,
reliability, input-output efficiency, and storage density. The most recent conclusion Ontario Hydro has
reached, on the basis of studies of large-scale energy-storage alternatives, is that:
... for early inclusion in Ontario Hydro’s generation program the most economical large-scale energy
storage alternative is pumped storage, either above ground or underground, and that no more than a
watching brief should be kept on alternative storage technologies. Among the alternatives studied
were: above-ground pumped storage; underground pumped storage; compressed air storage; feed-
water storage in a pressurized underground cavern; lead acid battery storage; and steam storageina
pressurized underground cavern.””
The implications of storage in the Ontario Hydro system are considered in Chapter 8. Briefly, it may be
said that pumped storage has quite a limited role to play in enhancing the operating flexibility of a
planned generation programme when faced with lower than expected load growth, because the storage
lead time is likely to equal that of a nuclear plant.
Summary and Conclusions
The mix of generating resources in an electric power system is affected by many factors, important
among which are cost, reliability and performance, operating characteristics, fuel requirement and
supply, lead time, and total-system considerations. An economic comparison of the conventional gener-
ation technologies used in Ontario indicates that, for base-load requirements, a new nuclear plant has a
substantial economic advantage over a new coal-fired plant. The cost of geological disposal of spent
nuclear fuel is quite uncertain but this cost has to increase by one order of magnitude to offset the
advantage of nuclear. Oil- and gas-fired plants are economic for peaking and reserve, but, if such
plants are used for decentralized applications in a co-generation mode, they may be economic at higher
capacity factors.
An analysis of the reliability data for Ontario Hydro’s thermal units shows that their performance has
been as expected and that the CANDU units should be able to maintain base-load capacity factors.
While CANDU units are not suitable for load-following, they can be shut down over weekends and can
be operated at reduced outputs overnight. Hydro’s CANDU units are planned for operation in this
century only at base-load capacity factors.
Ontario Hydro’s fuel supply appears reasonably secure. No problem in meeting the peaking and re-
serve requirements of residual oil- and gas-fired stations is foreseen. With respect to coal supplies,
Hydro has a problem of oversupply until the late 1980s. After that, the requirements, which will not be
much higher than current consumption levels, could be met by a combination of U.S. and western
Canadian supplies. Hydro’s current uranium contracts are adequate for the 30-year requirements of
about 5,400 MW of uncommitted nuclear capacity (after completion of the Darlington Generating
Station). Although the lead time of a major generating facility could be reduced from 13 years to eight
years by site-banking, there is no urgency about doing this. Sufficient time is available for public
participation and environmental assessment of any new site proposal by Hydro. In the longer term,
site-banking may play an important role in enhancing planning flexibility.
A major total-system consideration in the choice of a generating mix is the desirability of a diverse mix
to increase the system’s resilience and flexibility. In this regard it is important to note that although
economics is the major consideration in Ontario Hydro’s planned mix, operating limitations and fuel
diversity are also given consideration. For example, the economics suggest that about 60 per cent of
total system capacity should be nuclear, but Hydro’s planned share of nuclear is about 50 per cent. The
ability of Hydro’s fossil-fuelled stations to burn alternate fuels might not enhance the flexibility to any
significant extent, because, while it is possible to change existing coal-fired stations to burn oil or
natural gas (such a shift is considered unlikely), the converse is not true. What may be more important
is the ability of fossil-fuelled stations to burn biomass- or refuse-derived fuels. Operating flexibility,
particularly of CANDU stations when faced with lower than expected load growth, could be enhanced
by shorter lead time storage schemes to absorb surplus nuclear capacity during off-peak hours. How-
ever, this will not be possible with pumped storage (the most economical option), because its lead time is
likely to equal that of a nuclear plant.
Mix of Generating Resources 338
Figure 3.1 Economic Cost Comparison of a New Nuclear and a New Coal-Fired Generating Station Coming into Service in 1985
9,000
3,000
Accumulated lifetime discounted
cost in 1985 dollars per kilowatt
2,000
1,000
Sources: RCEPP and “Life Cycle Costs of Coal and Lifetime annual capacity factor (%)
Nuclear Generating Stations”, by S. Banerjee
and L. Waverman, July 1978: a study
commissioned by RCEPP.
Figure 3.2 Cost of Generating Unit Unreliability
5,000
4,000
A—Cost expressed in dollars per kilowatt of load-meeting capability
B — Cost expressed in dollars per kilowatt of nominal capacity
CANDU nuclear
3,000
Capital and 30-year discounted 0 & M
costs in 1985 dollars per kilowatt.
2,000
1,000
0 250 500 750 1,000 1,250
Unit size (MW)
Source: “Generation Planning Processes,” Ontario Hydro submission to RCEPP, May 1976, Exhibit 21.
34 The Electric Power System
Figure 33 Distribution of Hourly Capacity and Load for Ontario Hydro East System
120
| Hydraulic
ae Thermal
Percentage of annual peak demand.
0 20 40 60 80 100
Duration in per cent of time in the year
Note: The operating bands for hydroelectric capacity are rough approximations, because individual hydroelectric capacity factors will be distributed across the 0% to 95% range.
Sources: RCEPP and Ontario Hydro
Mix of Generating Resources 35
Figure 3.4 Pickering A Generating Station — Capability versus Age
100°
90
80
Capability
factor (%)
10
60
30
Years of service
Sources: Table 3.4 and “Generation Planning Processes,” Ontario Hydro submission to RCEPP, May 1976, Exhibit 21.
Figure 3.5 Efficient Operating Ranges of Conventional Power Sources
Generation
CANDU nuclear Fossil-steam Gas turbine Energy storage
f=p)
Average capacity
factor (%)
More likely economic range
Source: “Generation Planning Processes”, Ontario Hydro submission to RCEPP, May 1976, Exhibit 21
Less likely economic range es Non-economical or not technically feasible
36 The Electric Power System
0002
veld 6/61 OupAH o1e\UQ
A\ddns ajqepuedaq
“OUPAH OLED PUR Jq}9Y :Seanos
. 129)
66 86 L6 96 $6 16 4) 06 16 06 68 88 £8 98 68 08 £8 28 18 0861
0
g
BUIPIdIXA 40 BoUeYD % YG —Ajddns ajqeqoig
BUIPSdOXe 40 BOULYD %0G —Ajddns ajqepuadeg
ZA
uewa
puewed Zs “4
4 01
Fa
a
oa
oa
a
a”
7
poe
U0}8U!|Jeq Jaye UoIsuedxe ON Fe
pid $l
a
a
7
oa
ee S@UU0}
ee Jo Spuesnoy|
7
7
ao 02
ot
oo
7
7
o
7
2
7
at
a” G2
oa
oo
rg
oa
id
a
Oo 0¢
vote Ajddns ajqeqag
a
Pe
7
Ge
(f|UO g-dMH@ Pue Y-dMHg) Puewag pue Ajddng sa}eN Area} g'g “aunty
37
Mix of Generating Resources
\
”
WOU AB i AE a: #
i
ke bella
Lar aeg ee
CHAPTER FOUR
The Reliability of the Electric Power System
The reliability of a device or system is defined as the probability that it will fulfil its purpose adequately
for the intended period of time and under the specified operating conditions. In the case of an electric
power system, the purpose may be stated broadly as being to supply electricity to the customers with a
high degree of continuity. As will be noted later in this chapter, an electric power system is so complex
that it is neither feasible nor practical to express its reliability in terms of a single probability. The
probability of a malfunction of a component of a power system, such as a generating unit, a trans-
former, or a transmission line is nevertheless an important consideration in any evaluation of power
system reliability.
The causes of unreliability may be long-term or short-term in origin. Among the long-term factors that
may affect reliability are the lead times (in the order of a decade) that are required for the installation
of major new generation and transmission facilities and the uncertainty that surrounds the demand
for electricity over such a long period. A project may fall behind schedule due to strikes, delays in
equipment deliveries, mismanagement, or delays in obtaining government approval in matters that
have political or socio-environmental implications. Because of the length of the lead time, the chance of
offsetting such a delay by advancing the schedules of other projects is likely to be poor. Even if there is
no delay in the in-service date of the project, the actual demand for electricity may by then differ
significantly from the forecast demand, in level, in load shape, or in geographical distribution. Another
uncertainty in the long term has to do with the effective utilization of generating capacity. Actual
output may be constrained by changes in environmental regulations, by fuel shortages, or by droughts.
In day-to-day operations, the predominant source of unreliability is the unavoidable or unexpected
forcing out of service of part of the supply system. A generating unit may be forced out of service bya
component breakdown caused by wear and tear or by inherent defects. A transmission line may be
disrupted by lightning. Every generating unit must be removed from service periodically for preven-
tive maintenance, to reduce the incidence of unexpected failures. Another short-term consideration
that may cause a shortage of capacity is an increase in demand as a result of changes in the weather or
in the level of economic activity.
A utility’s customers require a degree of reliability that is considerably higher than the degree of
reliability of the individual components of the system. It is necessary to have spare components, and
spare capacity, so that the system can continue to supply users satisfactorily when some of its compo-
nents are unavailable, and to provide against the other contingencies that have been mentioned.
Additional capacity for generation is called reserve generating capacity, and for reasons of fuel econ-
omy it is used only when necessary. Reserve capacity is commonly referred to as “reserve margin”? it
represents the difference between the generating capacity and the peak demand, expressed as a
percentage of the peak demand. In Ontario Hydro’s system, for the purposes of reliability evaluation
and system planning, the peak demand is divided into two categories — primary peak and firm peak.
Firm peak is defined by Ontario Hydro as the peak demand that must be supplied to the customers with
a high degree of continuity. Ontario Hydro also offers its large industrial and commercial customers
“interruptible” power, which is supplied at a lower level of reliability but at cheaper rates. As the name
implies, this service may be interrupted at times when the system supply capability is unable to sustain
it. The primary peak demand is the sum of the firm peak demand and the interruptible power. The
generating reserve margin in Ontario Hydro’s system is usually expressed as a percentage of the firm
peak demand. A measure of reserve capacity for the bulk power transmission system is not available.
As will be discussed later, the reliability of a transmission system depends more on the behaviour of the
system asa whole than on the reliability of its individual elements.
Availability and Security
Since the electric supply system is made up of three major subsystems — generating stations, bulk
power transmission lines, and a complex distribution network — the reliability of supply to the cus-
tomer, for a known demand, depends on the reliability of these three components as well as on the
design and operation of the interconnected system.
Recently, Ontario Hydro undertook customer surveys to determine the costs imposed on customers by
The Reliability of the Electric Power System 39
supply outages of varying frequency and duration, and it is attempting to relate this information to the
design of the system. However, traditionally, the reliability of each of the major subsystems has been
assessed separately and on the basis of different performance criteria.
The traditional methods of reliability analysis assume that the reliabilty of each major subsystem must
be assessed from two aspects — “availability” and “security”. Availability refers to the performance of
the individual components of a system (generating units, transmission lines, transformers, circuit-
breakers). It is defined as the expected percentage of the total time during which the component is
required, that is, not out of service due to a fault, an equipment failure, or incorrect operation or
maintenance. Thus, when we say that the availability of a generating unit is 80 per cent, we mean that,
on the average, the unit will be unavailable for 20 per cent of the time. Component availability is
estimated from the outage records. These data are essential in evaluating subsystem reliability.
Security is aterm that is applied to an electric power subsystem or to the system as a whole. It is defined
as the ability of asystem to withstand transient disturbances and regain an acceptable operating state.
An acceptable operating state exists when the voltages and line loadings across the system are within
acceptable limits. After a major disturbance, which may be caused by a transmission line fault, the loss
of alarge generator, or asudden large change in the load on the system, the system enters what may be
called a dynamic state. While the system is in a dynamic state, the power flows and the voltages across
the system are changing continuously as the system seeks a new balance to satisfy the changed condi-
tions. If a satisfactory new balance can be achieved, the system is then said to be secure. If the power
swings increase in magnitude to such an extent that instability results, leading to the disconnection of
generators and lines by protective devices, and eventually perhaps to widespread and lengthy power
interruptions, the system is not secure. It is stressed that, although a system may be stable, it is not
secure unless it regains an acceptable operating state after the disturbance.
The massive northeastern black-out of 1965 is a classic example of an insecure system. An outage on
Ontario Hydro’s transmission lines running west from the Sir Adam Beck Generating Station at
Niagara Falls blocked the flow of power towards Hamilton and Toronto, blacking out southern Ontario.
The Niagara power took the only path open to it and rushed over the interconnections into New York.
This sudden surge of power caused a variety of overload protection relays to trip in a cascade that led to
a collapse of the entire New York system and the systems of the adjoining states.
Table 4.1 summarizes the availability and security aspects of reliability in the major subsystems of an
electric power system.
Table 4.1 Steady-State (Availability) and Dynamic (Security) Aspects of Power System Reliability
Steady-state aspect Dynamic aspect
Subsystem Generation Transmission Distribution Bulk power
(generation and transmission)
Typical failure causes Generator outages Line or transformer Line or transformer Multiple line outages
outages outages in a short time
Unexpected load Unexpected load Unexpected load Protection system failure
Resulting conditions Generation deficiency Line overload Line overload Cascading outages
Energy deficiency Low bus voltage Low bus voltage Instability
Service consequences System voltage reduction Low voltage (areawide) Low voltage (local) System blackout
Systemwide load reduction Areawide load reduction Local load reduction
(selective customer (selective customer (customer interruptions)
interruptions) interruptions)
Source: “EPRI Journal’, Electric Power Research Institute, Vol. 3, No. 10, December 1978, p. 10.
Availability and Security in the Generation Subsystem
As far as the reliability of a generating system is concerned, the availability of a generating unit is the
predominant factor. Availability accounts for both scheduled and unscheduled shut-downs of the unit.
Various indices are used by utilities to express generating unit availability. The following four are
appropriate for this discussion: the forced outage rate, the planned outage rate, the maintenance
outage rate, and the capability factor.
The “forced outage rate” (FOR) is the ratio of forced outage hours to the sum of operating hours and
forced outage hours. It is a measure of the incapability of a generating unit to produce energy due to
40 The Electric Power System
forced shut-downs. Forced outages are random in nature and require a unit to be derated or taken out
of service as quickly as possible.
The “planned outage rate” (POR) is defined as the fraction of a period, e.g., a year, when a generating
unit is out of service for planned maintenance. Planned maintenance includes major overhauls to
reduce the incidence of forced outages, and it may be postponed from one season to the next. It is
scheduled months in advance and is generally carried out annually during the season of low loads.
Maintenance outages, like the forced outages, are random in occurrence but they do not require a unit
to be derated or taken out of service immediately. They are generally scheduled for ‘‘safe” periods
when they will not interfere with the utility’s ability to supply the load fully, e.g., during nights or at
weekends. The “maintenance outage rate” (MOR) is the fraction of a period when a unit is on mainte-
nance outage.
The “capability factor” is a measure of the ability of a generating unit to deliver energy in the absence
of any problems or restraints external to the unit.! It is expressed mathematically as (1-FOR)(1-POR-
MOR). The capability varies with the type and size of the unit. Hydraulic units are the most reliable. The
average capability factor of Ontario Hydro’s hydraulic units is about 95 per cent. Large thermal gener-
ating units have mature capability factors in the range of 75 to 80 per cent.
The outages for a generating unit are generally highest during its initial years of service. With time
the unit “matures”, the outages decrease, and the capability increases to a mature value. As an exam-
ple, Ontario Hydro’s forecasts of the first-year and the mature capability factors of a500 MW CANDU
generating unit are 68 per cent and 80 per cent, respectively. The maturing period is in the order of
four to five years for large thermal units.
Security is not a prime concern in assessing the reliability of a generating system. Generating system
security is inherently high because (a) generating units are designed to limit the number of sudden
losses of generating capacity, and the design of the bulk power system is aimed at accommodating
losses that do occur, and (b) generating stations are designed to cope with sudden stresses that may be
imposed on them by problems in the bulk power transmission system. It should be noted, however, that
for certain disturbances and transmission line outages it may become necessary to “reject” generation
in order to improve the transient stability of the bulk power system. Generation rejection means
disconnecting one or more generating units from the bulk power system to reduce the excess power.
Hydraulic units are quite rugged and are therefore preferred for rejection. While the thermal units are
also designed to withstand the shocks caused by sudden disconnection, frequent rejection at full load
may cause damage and so reduce their availability.
Availability and Security in the Transmission Subsystem
The role of a transmission system may be stated, in general, as being to provide a link between the
generating stations and the system load. Since there are many generating stations and many load
centres, the link is in the form of a network. The availability of the elements of this network is largely a
function of their failure and repair rates. Individual transmission elements tend to have much higher
availability than generating units. This is particularly true of the Ontario Hydro system; in 1974, the
availability of its major transmission lines averaged 98.8 per cent.
However, from the viewpoint of the reliability of the bulk power transmission system, security rather
than the availability of transmission components has traditionally been the main consideration in
transmission planning in Ontario. Transmission planners historically have recognized that the trans-
mission system has the potential for producing complete system collapse — as was vividly demonstrated
in the northeast black-out of November 1965, and again in July 1977 in New York City. A bulk power
transmission system is a highly complex mechanism. Its behaviour under abnormal conditions is corre-
spondingly difficult to predict. For example, a transmission line may be subject to different types of
faults (e.g., lightning or physical damage causing a short circuit) at a variety of locations along its
length. The dynamic behaviour of the whole transmission system will depend on the nature of the fault,
its location, and the events immediately after its occurrence, namely, how rapidly the faulty line is
removed from service by automatic protection and control devices, whether an immediate automatic
attempt is made to restore the line to service, and whether or not this attempt is successful.
Because of a transmission system’s complexity, no quantitative index of its security is available. The
transmission planners usually carry out simulation studies to determine a system’s ability to survive
certain postulated events (events that are severe but credible) without causing cascading failures and
The Reliability of the Electric Power System 41
system collapse. If the system can survive these events, it is expected to survive any others. The selec-
tion of these postulated contingencies requires the exercise of mature judgement based on broad
experience. This judgement is now being augmented by computer-aided security assessment.
Availability and Security in the Distribution Subsystem
Of principal interest at the distribution level is the availability of components and the continuity of
service at the customer’s premises. An outage of a single component of the distribution system gener-
ally leads to a local interruption of power of the sort that is responsible for the greater part of the
customer interruption time. The outage may be caused by equipment malfunction or by some external
event such as lightning, a storm, or the collision of a heavy vehicle with a distribution pole. Prolonged
power black-out may result if, after a lightning strike on a distribution feeder, the circuit-breaker at
the distribution station fails to close again.
The security of the distribution system is of little consequence, since disturbances on the distribution
network generally have little impact on the bulk power generation and transmission system.
Reliability Evaluation
Reliability evaluation deals with the question: What increase in reliability is associated with adding a
piece of equipment to the system, or, What is the reliability of the system with and without the addi-
tional equipment? Power system planners have been trying for decades to develop ways of producing
consistent yardsticks of power system reliability. Without such yardsticks, reliability evaluation has to
depend heavily on experience and judgement. The reliability levels of alternative plans with different
equipment sizes or equipment availabilities cannot be compared accurately; the effect of changing load
patterns on system reliability cannot be accurately assessed. Without yardsticks of reliability, it is
difficult to take these complexities and uncertainties into account in the planning process. It is, also,
difficult to quantify the conclusions — to state with any certainty why new facilities are needed and
when they should be added.
Ideally, the methods used for evaluating reliability should produce measures, or indices, of at least three
significant consequences of a failure to provide electrical service. These are:
e the magnitude, interms of power and energy, of the electrical load that fails to be supplied in each
occurrence of failure
e the frequency with which the load fails to be supplied
e theduration of each occurrence.
Allthree measures are important because the benefits of reliability, or the customers’ perceptions of the
cost of unreliability, depend on the magnitude, frequency, and duration of an outage. In the opinion of
the electricity industry, a quantitative expression of the three consequences of a failure to supply is an
adequate measure of the reliability of an electric power system.
Since the purpose of an electric power system is to provide its customers with a reliable supply, the three
indices of service quality must be known at the customer end. This entails the evaluation of the reliabil-
ity of each subsystem — generation, transmission, and distribution — using a common methodology for
all three so that the reliability of the subsystems can be combined to give the customer an index of
overall reliability. At present, the reliability of each subsystem is evaluated separately, and a different
methodology is used in each case. Workable techniques for measuring the reliability of a total power
system, as viewed from acustomer’s premises, do not exist.
What follows is an outline of the techniques commonly used in evaluating the reliability of the genera-
tion, transmission, and distribution subsystems, and of efforts that have been made by Ontario Hydro
and other utilities to improve these techniques and integrate them so as to produce a measure of
reliability from the customer’s viewpoint.
Generation Reliability Evaluation
The reliability of the generating system has long been a matter of primary interest to the electricity
utility industry. Probability techniques for estimating the generating reserve requirements have
been in use for many years. The basic concern has been the provision of sufficient reserve capacity to
assure within reason that available capacity will not fall short of load requirements at any time due to
forced outages of generators, generator maintenance requirements, or load-forecast uncertainties.
Several papers have been published on probability methods for the evaluation of generation reliability,
42 The Electric Power System
and the developments that led to the techniques now in use are interesting. The application of probabil-
ity techniques in this field was discussed as early as 1933, but it was not until 1947 that the first major
papers were published. These papers advanced some of the concepts on which the methods in use today
are based. These methods, with some modifications, are generally known as the “‘loss of load approach”’
and the “frequency and duration of outage approach”.
Although these techniques are available, many of the smaller electricity utilities still employ genera-
tion reliability standards that are based on deterministic methods dating back to the early days of the
industry. These include the “per-cent reserve” method, in which generating capacity is added to the
system to maintain the reserve at a given percentage value of the annual peak load, and methods
expressing reserve in terms of the capacity of the largest units in the system (e.g., “reserve equal to 150
per cent of the capacity of the largest unit” or “reserve equal to the capacity of the two largest units”).
Deterministic rules are arrived at by judgements based on experience and historical observation.
Deterministic methods do not provide for any quantitative assessment of the reliability of various
generation planning alternatives. Yet they are in widespread use, probably because of their simplicity.
A survey conducted by the Canadian Electrical Association in 1974 indicated that only four of the 13
Canadian electricity utilities canvassed used probability methods to calculate generation reliability.”
The other nine used “per cent reserve”, the size of the largest unit, or a combination of these two as the
criterion for planning generation. It must be noted, however, that most systems using the determinis-
tic rules undertake probability studies periodically to confirm the adequacy of these rules.
On the other hand, the methods based on probability analysis can take into account the effect on
reliability of such factors as the size, type, and number of generating units, the availability of the units,
interconnections among systems, the shape of the load curve, and the uncertainty of load forecasts.
Thus, probability methods can provide a consistent basis for comparing various planning alternatives,
and they are far more useful to the system planner than simple deterministic rules.
The Loss-of-Load Probability Method
The probability technique most widely used by the electricity utilities for estimating the reliability of
their generating systems is the loss-of-load probability (LOLP) method, or, as it has more recently and
perhaps more correctly been termed, the loss-of-load expectation method. This method computes the
expected (long-term average) number of days per year on which the available generating capacity is
not sufficient to supply all of the daily peak load. The basic assumptions of this method may be summa-
rized as follows:
e Transmission and distribution limitations are neglected and the total generation and load are
assumed to be concentrated at the same point in the system.
e Generating units are assumed to be independent of each other. That is, it is assumed that the
outage of one unit will not affect the operation of the others.
e Loadisrepresented by the daily peak loads. Itisassumed that if there is enough capacity tosupply
the daily peak, the rest of the daily load can be supplied.
e At the time of the daily peak, all of the generating capacity that is not on forced or scheduled
outage is assumed to be available to supply load.
e Itisgenerally assumed that generating capability will not be limited by fuel shortages. However,
it is possible to analyse the effects of such limitations, if necessary.
Basically, a probability distribution of the amount of generating capacity that is likely to be rendered
unavailable by an outage is obtained from data on the sizes and forecast outage rates of the generating
units comprising the system. The capacity of the equipment that is out of service is considered to be a
load on the system and is therefore added to the daily peak loads by combining the distribution of the
capacity on outage with the distribution of the daily peak loads over a given period, for example, a
month. From the resulting distribution of ‘equivalent load”, it is possible to determine the probability
that the equivalent load is higher than the peak installed capacity during the chosen period. This
probability is the LOLP for that period. The LOLP can be translated into the expected number of loss-
of-load days by multiplying the probability by the number of days in the period. The analysis can be
repeated for other periods to obtain the loss-of-load days per year.
Among the utilities using the LOLP method for planning generation, acommonly accepted value of the
reliability index is 0.1 days per year or one day in 10 years. This means that sufficient generation
capacity will be installed that the expected number of loss-of-load days in a given future year will be no
more than 0.1. In the past, Ontario Hydro has used this value as its target for generation reliability.
The Reliability of the Electric Power System 438
Hydro’s LOLP method is similar in principle to, but different in details from, the LOLP computations
made by other North American utilities. Hydro’s technique calculates the LOLP on a month-by-month
basis by making appropriate allowance for the generating capacity that will be on planned mainte-
nance. As far as the load model is concerned, Hydro considers only the firm 20-minute peaks on the
working days (the peaks on weekends and holidays are usually low enough that the chance of insuffi-
cient generating capacity at such times is negligible). This amounts to about 240 working days per year
and the corresponding target LOLP is 1/2,400. Hydro schedules the maintenance of its generating
capacity in such a way that the LOLP in each month is in the neighbourhood of 1/2,400, the so-called
“equal risk criterion”. Very little maintenance is carried out during December and January. Thus,
Hydro’s reliability calculations are, in effect, based on a policy of providing reserve against any forced
outages of generating units during these months. The reserve against the planned maintenance of
generating capacity is provided, in effect, by the lower loads that occur during the off-winter months
(see Figure 2.5 in Chapter 2).
The reserve margin required for a given LOLP, such as 1/2,400, depends very much on the forced
outage rate and on the size and number of generating units in the system. For the Ontario Hydro
system, the reserve requirements have been in the 25 to 30 per cent range for an LOLP of 1/2,400,
taking no account of the assistance available from the interconnections with neighbouring systems.
Figure 4.1 gives an indication of the sensitivity of the LOLP in Hydro’s system in response to varia-
tions in the generating reserve margin. The reference value for the reserve margin is that which
results in an LOLP of 1/2,400. The estimated variation in Figure 4.1 is derived from Hydro’s long-
range forecasts LRF 48 and LRF 484A. As may be seen, the computed LOLP changes exponentially with
the reserve margin, expressed as a percentage of the peak load. A 5 per cent variation in reserve
margin changes the LOLP by more than an order of magnitude.
While reliability depends on a multitude of factors, significant among factors that are amenable to the
widely used probability methods are: unit forced outage rate; unit size; the availability of assistance
from neighbouring systems; and the uncertainty about the size of the load and about the in-service
dates of generating units. The following is a discussion of the effect of these factors on the reserve
margin requirements for the Ontario Hydrosystem.
Figure 4.2 indicates in a general way the reserve generation requirements associated with generating
units of different sizes and forced outage rates. It illustrates this by considering the effect of adding a
number of identical generating units to Ontario Hydro’s existing, under construction, and committed
system (with a generating capacity of about 33,000 MW) for the purpose of supplying an additional
load of 5,000 MW witha specified LOLP of 1/2,400. The reserve is the difference between the additional
generating capacity and the incremental load, expressed as a percentage of the incremental load. It is
natural to expect system reliability to deteriorate, or reserve requirements to increase, with an in-
crease in the forced outage rates of units, other factors remaining constant. Figure 4.2 shows that, if a
5,000 MW increment in load is to be supplied by 750 MW units with a forced outage rate of 10 per cent,
the reserve requirement is about 21 per cent, that is, a total additional generating capacity of 6,050
MW. If the forced outage rate is 7.5 per cent, it is 14 per cent; and if the forced outage rate is 12.5 per
cent, itis 27 percent.
As for the effect of unit size on generation reliability, a system containing twenty 100 MW unitsis more
reliable than one containing ten 200 MW units with the same forced outage rates. As an illustration,
consider one 100 MW and two 50 MW units, each with a 10 per cent probability of failure. With the 100
MW unit, the probability of aloss of 100 MW of generating capacity is 10 per cent, compared with 1 per
cent in the case of the two 50 MW units. Thus, the reliability of a generating system deteriorates as the
size of the units in relation to total system capacity increases, or as the number of units decreases.
Figure 4.2 shows that the reserve requirements associated with the 750 MW units with a forced outage
rate of 12 per cent are 26 per cent of the incremental load; but, with 1,250 MW units of the same forced
outage rate, they are 37 per cent of the incremental load.
Interconnections between systems improve the overall level of system reliability. In the case of two
interconnected systems, the reliability of both systems is enhanced by the diversity in the occurrence of
outages of their generating units. That is, there is some chance that, when one system is unable to
supply its load completely, the other may, at that time, have surplus generation available that can be
transferred across the interconnection to supply the first system’s load. At other times, the situation
may be the reverse, with the second system receiving assistance from the first. This kind of assistance
allows each of the interconnected systems to operate with less reserve than it would require under the
44 The Electric Power System
Fig. 4.1: p. 54
Fig. 4.2: p. 54
Fig. 4.3: p. 55
conditions of isolated operation. This possibility is illustrated in Figure 4.3. The figure shows the results
of a hypothetical situation in which Ontario Hydro’s existing East System, along with the Pickering B
generating station (with a total generating capacity of about 26,300 MW), has been interconnected
with another, identical, system. With an interconnection capacity of 2,000 MW, each system can reduce
its reserve requirement by about 1,400 MW (compared with the situation when they are not intercon-
nected) and still maintain the same level of reliability with an LOLP of 1/2,400. Figure 4.3 also shows
the corresponding results for two lower levels of generation reliability. Another interpretation of
Figure 4.3 is that, if the reserve margin in each of the interconnected systems is held constant, then the
reliability of each system increases with interconnections. Interconnections can also be beneficial, in
the matter of reserve requirements, if there is diversity in the loads of the interconnected systems.
Such diversity may be in the form of a seasonal climatic diversity, e.g., as between New York and
Ontario, or in the form of a daily time-zone diversity, e.g., as between Manitoba and Ontario. The
extent of the benefits of diversity between two systems depends on the amount of diversity, institu-
tional arrangements, and the maintenance schedule for generating units. Interconnections also tend
to reduce the shocks caused by major contingencies and to facilitate economizing transactions. These
issues are discussed in a later chapter. The discussion of the effect of interconnections on system
reliability will not be complete without mention of a major disadvantage of interconnections: intercon-
nections make the operation of the power system more complex, necessitating constant co-ordination
among member systems.
The effect on generation reliability of the uncertainty concerning the size of loads and the in-service
dates of units depends on the nature of the uncertainty. Most load forecasts deal with the most probable
load in a given future year, so that there is equal probability that the load will be greater or smaller than
the forecast size. It is evident from Figure 4.1 that the reliability decreases more and more rapidly as
the reserve margin is reduced. Thus, the expected value of reliability, if the size of the load is uncertain,
will generally be lower than the value corresponding to the most probable load. The uncertainty of the
in-service dates for new units may be treated in the same manner as the uncertainty about the size of
the loads; delays in in-service dates are equivalent to higher-than-expected loads, and units coming
into service before their expected dates tend to increase the reserve margin, as do lower-than-forecast
loads.
Variations in the Use of the LOLP Method
While the LOLP method has gained reasonably wide acceptance, variations in the way in which elec-
tricity utilities apply it, and dissimilarities in the utilities’ system characteristics and configurations,
can lead to appreciable differences in reliability levels and reserve margins, even though the basic
reliability criteria are essentially the same. Asasimple illustration, Ontario Hydro, Hydro-Québec, and
Manitoba Hydro all use a reliability criterion based on an LOLP of ‘one day in 10 years” for determin-
ing generation reserve capacity.® However, Ontario Hydro incorporates only working-day peak loads
over a month in its LOLP model, whereas Hydro-Québec uses a monthly load-duration curve represent-
ing loads on both weekdays and weekend days. Thus, the risk index of one day in 10 years means an
LOLP of 1/2,400 to Ontario Hydro but an LOLP of 1/3,650 to Hydro-Québec. On the other hand,
Manitoba Hydro’s model is based on only the weekday peak loads in the month of January, but the risk
index in days per year is obtained by multiplying the LOLP by 365. And there are other differences
between the utilities with respect to the assumptions they make when evaluating reliability by the
LOLP method. Most utilities do not consider interconnection assistance in their LOLP computation.
But those who do this use a lower risk index; for example, Manitoba Hydro uses a risk index of 0.003
days per year when considering interconnection assistance. Some utilities recognize the uncertainty of
load forecasts in their computations and others do not.
It is clear, then, that an LOLP risk index of 1 day in 10 years may result in different generation
reliability standards. For a particular utility, the strength of the LOLP technique may lie in its relative
simplicity and in its ability to compare the reliability of alternative generation plans by analysing the
effects of unit size and the forced outage rates of units, the uncertainty of load forecasts, and the
availability of assistance over interconnections. This technique falls far short of providing an adequate
measure of reliability for a given generation plan in the matter of the size, duration, and frequency of
interruptions to the supply. Since the LOLP technique uses only the probability of an outage of a
generating unit, itcan make no prediction about the frequency of interruption, that is, the “how often”
aspect of reliability. (It should be noted that the LOLP risk index of 1 day in 10 years must not be looked
upon as the expected frequency of capacity shortages. Such an index only expresses the probability in
The Reliability of the Electric Power System 45
lay language.) Nor does the LOLP method provide any measure of the magnitude of an interruption in
terms of power or energy, or of its duration.
The Frequency and Duration Method
These shortcomings of the LOLP technique led to the development of the frequency-and-duration-of-
outages (F&D) method. This method recognizes the failure and repair rates of generating units, and
not just the forced outage probability, and gives a more complete and useful measure of generation
reliability. The relationship between the three quantities is illustrated by the following example. If a
generating unit fails, on the average, every 100 days (0.01 failures per day) and it takes 25 days, on the
average, to repair it (0.04 repairs per day), then the forced outage probability is 0.2. The F&D model
consists basically of two parts: a capacity model and a load model. From the data on the sizes, the failure
rates, and the repair rates of generating units, the capacity model can calculate the frequency and
duration of each capacity-on-outage state. The load model calculates the frequency and duration of
assumed load levels from the daily load cycles over the period concerned. The two models are then
combined to produce the desired index of reliability in terms of the frequency, the duration, and the
magnitude of capacity shortages.
In response to recommendation III-19 of the Select Committee of the Ontario Legislature in June 1976,
Ontario Hydro has developed a programme based on the F&D method to evaluate the reliability of its
generating system. This programme was used by Hydro in its System Expansion Program Reassess-
ment (SEPR) study to complement the results of the LOLP technique. Table 4.2 provides a comparative
assessment of the features of these two techniques as used in the SEPR study. The methodology used in
SEPR toassess reliability will be discussed later.
Table 4.2 Features of the Loss-of-Load Probability (LOLP) and Duration (F&D) Reliability Computations of Ontario Hydro
Feature LOLP F&D
Load model
Peak loads
— Weekdays December only Yes
— Weekends No Yes
Off-peak loads No Yes
Load forecast uncertainty No Yes
Interruptible loads Yes Yes
Managed loads Peak load reduced by Simulation of effects for
a fixed amount daily peak modification
Generation model
Existing plant Yes Yes
Proposed plant Yes Yes
In-service date uncertainty No Yes
Output during commissioning No Yes
Variation in thermal unit outputs
due to normal temperature changes Yes Yes
Reduction in output due to
known government regulations Yes Yes
Variation in output from Sir Adam Beck
G.S. due to normal river-flow variations Yes Yes
Energy production limits at
other hydraulic stations No Yes
Planned outages Yes Yes
Maintenance outages No Yes
Forced outages and derating Yes Yes
Immaturity effects on outage factors Yes Yes
Uncertainty in outage factors No No
Common cause failures No No
Malicious damage, sabotage No No
Shortage of critical materials No No
Strikes No No
Failure in delivery of purchased firm power No No
Need to maintain operating reserve No No
Source: Ontario Hydro, System Expansion Program Reassessment Study, Second Interim Report, November 1978, p. 11.
46 The Electric Power System
Transmission Reliability Evaluation
The techniques available for evaluating the reliability of the bulk power transmission network are not
as well developed as those for evaluating generation reliability. This is primarily because of the diff-
culty of defining what is and what is not a failure in a network system and in understanding the
complicated and interrelated nature of events leading to a failure. In the generation sector, a failure is
said to occur when “available generation is less than load”, and this is usually caused by overlapping
outages of several generating units. Similarly, in the distribution sector, a failure is defined as an
interruption of customer supply caused by the outage of a component of the distribution system, such as
a distribution feeder. A similar definition can be used for the transmission sector, and, in fact, existing
transmission reliability models are based on this definition. But since transmission, in general, isa link
between generating stations and distribution centres, the bulk power transmission design philosophy
has traditionally been to avoid “widespread”, ‘‘cascading”, and ‘‘uncontrolled” power interruptions.
Such interruptions usually are not the result of overlapping outages of one or more components, but of a
chain of events having some interrelationship.
The ability of a system to avoid widespread black-outs is established by security analysis. Transmission
planners carry out a mathematical simulation of the bulk power system on a computer. The system is
subjected to selected fault conditions, fault-clearing sequences, and other contingencies that are severe
but nonetheless credible, and then it is determined whether the resulting modelled transmission per-
formance (voltage levels, line loadings, and the like) is within acceptable limits. The disturbances
chosen for simulation are based on the experience and judgement of the planner. Sufficient transmis-
sion is then planned to maintain system performance despite the occurrence of the more likely outages
(e.g., failure of a single line). However, some deterioration in system performance may be tolerated for
less likely circumstances, such as multiple line outages or the loss of an entire power plant. The idea
behind this approach to transmission planning is to reduce the chances of widespread black-out, but, if
it occurs, to limit its impact in terms of the number of people affected and to reduce the time it will take
to restore service.
Ontario Hydro, along with the New Brunswick Electric Power Commission and 19 utilities in the
northeastern U.S., is a member of the Northeast Power Co-ordinating Council (NPCC), which was set
up in the wake of the massive black-out of November 9, 1965. The council acts as a central co-ordinating
agency for the planning and operation of the interconnected utilities’ power systems to ensure ade-
quate reliability of service to the customers of each system. The NPCC has formulated guidelines for
the design and operation of the interconnected systems and these have been adopted by the member
systems. The three key principles embodied in the guidelines are:
e There should be adequate transmission capacity to ensure that, in the event of the failure of a
system element such as a generator, a transmission circuit, a transformer, or a circuit-breaker, no
cascading outages or major power interruption will occur.
e Thesystem should be operated within limits so that the loss of one element of the system will not
precipitate cascading outages.
e Plansshould be madeto minimize the size and duration of outages resulting from operating error
or from the loss of multiple facilities such as a four-unit generating station, all the circuits on a
transmission line right of way, or a major load.
The security methods provide only a qualitative measure of reliability; they do not provide quantitative
yardsticks of the magnitude, frequency, and duration of interruptions. The methods that have been
proposed for quantitative reliability evaluation are based only on the steady-state aspects (see Table
4.1). Because a transmission network has both “series” and “parallel” connections of components,
determining the cause of an interruption in service depends on the identification of a particular
component. Therefore, security methods must consider all possible combinations of component ou-
tages. The proposed methods include the following basic steps:
e Select an operating state of the bulk power system that involves the outage of one or more
elements.
Dispatch power tosatisfy the load.
Determine the power flow on each transmission link.
Check for overloads and low-voltage conditions.
If alink isoverloaded, itis assumed that the outage will cause a failure.
Calculate the amount by which the load has to be reduced at various load points.
By repeating this procedure for each state of the system, and analysing the data on the frequency and
Sr nn nn ne ee UtttttIIEIIE IIIS
The Reliability of the Electric Power System 47
duration characteristics of each state, it is possible to obtain reliability indices at specific system load
points, in stated system areas, or for the entire system. But the number of possible states for a practical
system may be extremely large (2" for a system with N components). In order to keep computational
requirements within reasonable limits, procedures must be devised for selective evaluation of system
states that will give results of acceptable accuracy. These procedures include ““Monte Carlo” simulation,
in which only a randomly selected sample of states is examined, and the ranking of outage states of the
system according to their probability of occurrence and their adverse impact on the system.
Toamass sufficient statistical data to provide a consistent basis for probabilistic estimates of the failure
rates and types of failure of the great many components that are involved is a challenging task. To
evaluate reliability under dynamic operating conditions, transmission network analysis must deal
with system stability under specified fault, or short-circuit, conditions. The models must therefore
simulate realistic responses, such as cascading, the activation of the protection apparatus, and a load
curtailment as dictated by various operating policies. Attempts must be made to include the effects of
planned outages of generating units and transmission lines, of daily, weekly, and seasonal variation of
loads at different points, and of multiple outages involving some common cause. Integrating these
several model capabilities into realistic and practical procedures that give a quantitative measure of
reliability is a challenge to the electricity utility industry, and considerable attention is being devoted
to it.
Ontario Hydro is involved in a computer programme called PCAP, which has been developed jointly by
Power Technologies Inc. and the NPCC in a move to advance the state of the art in reliability analysis.
PCAP embodies the basic ideas outlined in the preceding discussion. An important feature of the
programme is its ability to select automatically and subsequently test outage states that are likely to
result in system failure. Thus, the number of cases that must be analysed is greatly reduced. Although
PCAP determines the adequacy of the bulk power supply system under static conditions, NPCC hopes
that, through this programme, significant system contingencies that may escape detection with com-
mon planning methods may eventually be foreseen.
Distribution Reliability Evaluation
In the distribution sector, a failure is defined simply as an interruption of customer service due to an
outage in the distribution system. Probability techniques are available to evaluate distribution reliabil-
ity but their application has not been as extensive as in the generation sector. Moreover, the availability
of data on the outage rates of distribution equipment has been limited.
The traditional way of evaluating reliability in the distribution system has been to measure the reli-
ability actually experienced by various customers and to improve it, where needed, with better equip-
ment, better maintenance practices, or system reinforcements. Utilities maintain statistical records of
the outage data on equipment, by the type of equipment, its manufacturer, and its age. These data
serve as a guide for future purchases and help to determine the best time for retiring each piece of
equipment. The utilities often employ quantitative methods of reliability analysis when they wish to
modify their existing designs, for example, when there is a move to a higher primary distribution
voltage. Such a move would increase the number of customers that could be served per circuit, but at the
same time it might increase the likelihood of power interruption because of the need to use longer
feeders. Reliability analysis can help ensure that the move to a higher voltage will not reduce the
reliability of supply. Another example of a situation in which modifications in existing designs are
required is the conversion of overhead distribution to underground.
In general, the outage of a single component in the distribution system will cause an interruption of
supply to some customers. Because of this, the techniques for evaluating distribution reliability are
geared to single-component failures. The duration of an interruption caused by such a failure depends
on many factors. A piece of equipment may be permanently damaged and have to be repaired or
replaced. It may be rendered inoperative temporarily by some external cause such as lightning, and the
interruption of supply will then be momentary if the protective devices work properly. But if a protec-
tive device such as a circuit-breaker fails to re-close after a lightning strike, the interruption may be
prolonged.
In Ontario, most of the distribution systems are owned and operated by public utility commissions
(PUCs). The PUCs do not favour a reduction in the existing distribution reliability standards. Their
main argument is that they are the principal targets for complaints by customers in the event of a
supply interruption, regardless of whether the outage is in the distribution system or the bulk power
48 The Electric Power System
system. It is natural for the commissioners, managers, and staff of the PUCs to want to avoid public
criticism. Many customers are served from the distribution centres by a single feeder, a mode of service
with minimal reliability. Only customers whose cost of interruptions is very high are provided with
multiple feeders from alternate supply points. Certain industries, hospitals, large shopping centres,
commercial buildings, and large schools fall into this category.
Evaluation of the Reliability of Supply to the Customer
As was stated earlier, practical methods do not exist at present for evaluating reliability of supply to
customers on the basis of a quantitative evaluation of the reliability of the three subsystems of the total
power system. Attempts are being made by the power industry to integrate the models of generation
reliability and bulk power transmission reliability. The PCAP programme of Ontario Hydro is an
example. Depending on the success and acceptance of such methods, distribution reliability models
may be incorporated with them to provide measures of reliability to the customers.
Reliability of supply to the customers may be measured on the basis of past performance, by recording
the number of customers interrupted, the frequencies and durations of the interruptions, and the
amounts of load not supplied as a result of the interruptions. Data of this type may be used to predict
future performance. The recording of past performance also provides a measure of the relative past
contributions of the major subsystems - generation, bulk power transmission, and distribution - to
reliability of supply. Data of this type indicate that, in Ontario, loss of supply at the customer level has
occurred most frequently as a result of failures within the distribution system. Failures in the trans-
mission system appear to have had a relatively minor effect on reliability of supply to the customer, and
generation failures appear to have had little or no effect. Ontario Hydro customers have not suffered an
interruption of supply as the result of generation shortfall for many years. During the Commission’s
public information hearings it was pointed out that this is true of the last 25 years or so. There have
been occasions, however, when Ontario Hydro had to make emergency purchases of power from the
U.S. due to generation deficiency. This is particularly true of the late 1960s, when Hydro had installed
generation reserves of less than 10 per cent. In December 1976, some industrial interruptible loads
were cut because of a combination of several severe contingencies: abnormally cold weather pushed the
peak demand to unusually high levels; power was locked in at the Lennox Generating Station due to
transmission bottlenecks; and several units of the Nanticoke Generating Station were forced out of
service. The severity of the situation was reduced considerably by emergency purchases from neigh-
bouring systems, proving beyond doubt the benefits of interconnections.
An indication of the reliability of Ontario Hydro’s bulk power supply system (generation and trans-
mission) can be obtained by considering the average annual number of interruptions and the average
total duration of interruptions experienced per year at each delivery point of the bulk power system.
The data for the period 1970-75 indicate that, on the average, each delivery point suffered 1.85 inter-
ruptions per year and the total duration of the interruptions at each delivery point was 21.3 minutes per
year. Translated into an index of the availability of supply, this gives a value of 99.996 per cent. These
figures may be compared with the data on the supply of electricity to some of Ontario Hydro’s 700,000 or
so rural customers. Data from rural customers in nine areas of the province indicate that each customer
experienced, on the average, 2.88 interruptions per year, with a total duration of 182 minutes per year,
giving an availability index of 99.9654 per cent. Thus, in terms of the duration of interruptions, the
contribution of the failures in the bulk power system is small. The frequency of interruptions cannot be
compared properly because the figure for rural customers does not include brief interruptions lasting
less than a minute.
The Costs and Benefits of Reliability
Since reliability is an important characteristic of an electric power system, the question of how much
reliability, that is, how much generation reserve and how many transmission lines or other system
elements should be installed, is central in the designing of a system. In the operation of a power system,
the transmission lines are usually not loaded to their maximum physical capability. Thus, when a
generating unit has to be taken out of service due to an outage, power can generally be moved from
areas of surplus generation to make up the deficiency. To a certain extent, then, it is economically more
efficient if an electricity utility has reserve generating capacity in some parts of its system than it is if
each customer has to either buy his own back-up or storage device or suffer due to interruptions of
supply. For planning purposes, the question is how to determine the point at which the cost of extra
The Reliability of the Electric Power System 49
generating capacity (and hence reliability) created by the utility exceeds the costs incurred by the
customer as a result of power interruptions. The cost to the utility may be computed from system
planning studies, but the costs incurred by customers are extremely difficult to estimate. Some custom-
ers, such as an average residential customer, might not be greatly inconvenienced in good weather bya
loss of power for one hour. Others, such as a continuous process textile plant, might be severely affected
by a shut-down of only a few seconds. Thus, the costs of an outage will depend significantly on the time
of year, the type of customer affected, and the magnitude and duration of the outage.
Until recently, the electricity utilities in North America made no direct attempt to balance these costs.
However, there are a number of cases of increased reliability being purchased by a customer. For
example, a hospital may have stand-by generating equipment for back-up, and a particularly sensitive
industrial plant may have an extra transmission line in case of a line outage. Historically, in North
America, new fossil-fuelled generating units have been more efficient than the old ones, and so have
had lower fuel costs per unit of energy produced. Moreover, the economies of scale in the new, larger
plants have meant less capital cost per unit of capacity. Thus, with a growing demand for energy, the
incremental cost of capacity at the peak period was below the average cost and so the average cost was
dropping over the years. Under these circumstances, an approximate measure of the past reliability
that had been acceptable to customers was used for planning purposes.
Recent approaches to choosing an appropriate level of reliability for electric power systems attempt to
balance the benefits accruing to the customers and to society from reliability with the costs incurred by
the utility (which are, of course, passed on to the customer). The benefits to the customer and to society
are usually taken to be costs that can be avoided because of the existence of redundant units. The net
benefits are then the benefits accruing to the customers and to society, minus the costs incurred by the
utility in providing a given level of reliability. The “best” (or optimal) level of reliability is that which
maximizes the net benefits. In practice, utilities often use another, equivalent, formulation in which the
optimum reliability level is that which minimizes the sum of the direct costs incurred by the customer
and the indirect costs incurred by society at a given level of reliability and the costs incurred by the
utility in maintaining that level of reliability. Moreover, provided that the cost functions of the utility
and the customers are convex, the formulation may be further simplified to one of equating the incre-
mental cost to the utility of ensuring reliability with the incremental costs incurred by the customers
and by society as a result of interruptions.
Presentations to the Commission by Ontario Hydro indicate that Hydro is improving its criteria for
selecting reliability levels. The new methodology, which has been described in Ontario Hydro’s System
Expansion Program Reassessment (SEPR) study, balances the estimates of the costs incurred by
customers and society as a result of outages with the costs incurred by the utility in achieving a given
level of reliability.* For several planned reserve margins, it basically combines the expected frequency
and duration of outages of different magnitudes with the results of some user surveys of the costs to
customers of each type of outage. Then these results are combined with an econometric model to
produce estimates of the total economic cost of interruptions. The cost of a given measure of reliability
(SEPR uses generation reserve margin) is the sum of the direct costs to the customer of interruptions
caused by outages, the indirect costs to the province (such as a reduction in the gross provincial product),
and the costs to the utility for generation, transmission, and distribution. These costs are calculated
annually and the sum of their present value is defined as the total economic costs (to the society)
associated with a given set of assumptions concerning the system expansion programme and the
demand. In this work, most analyses assume that all generation and load are located at a single point, so
transmission and distribution are studied in much less detail.
This new approach, balancing the costs of outages incurred directly by the customer with the costs toa
utility of maintaining a given reliability level is a significant improvement on the loss-of-load-proba-
bility (LOLP) method, which used a target level based on historical considerations. For several years,
Ontario Hydro has been computing the frequency and duration of generation system outages, but the
F&D results were used to assess the performance of the system, not to design it. The application of the
F&D technique to the design of the system necessitated three further steps. First, the accuracy and
detail of the F&D computer programme were improved. Second, consumer surveys were undertaken to
supply some estimate of the customers’ perceptions of the costs incurred during outages of various
durations (e.g., very short, one hour, two hours, etc.). Third, a methodology was developed to combine the
F&D results, the historical transmission and distribution outage experience, and the consumer survey
results into estimates of the direct costs of unreliability suffered by Ontario Hydro customers and the
50 The Electric Power System
Fig. 4.4: p. 55
indirect economic costs suffered by the province. These costs were then combined with Ontario Hydro’s
expenditures to provide an estimate of the “total economic cost”. Then the minimum total economic cost
was used as the design criterion for system reliability.
We will consider, first, how the reliability of the total system is estimated in the SEPR study. Then we
will examine how the customer losses were estimated. Finally we will see how the total costs to the
system were obtained, so as to arrive at a minimum cost level of reliability.
The reliability of the generation system is evaluated by Ontario Hydro’s F&D programme, as described
earlier. The programme computes the expected frequency and duration of generation outages, and the
expected amount of energy not supplied on demand during the year because of generation deficiency.
These indices are used as measures of generation reliability. Since a transmission system’s reliability
is, in practice, much more complex and hence more difficult to assess than a generation system’s, it is
assumed in the study that the transmission system’s historical performance will continue into the
future. For the “reliability balance” calculations, it is assumed that the energy unsupplied due to
transmission failures will remain a constant fraction of peak demand, and that the fraction of outages
of a given duration will remain constant as the system grows. The effect of these two assumptions is
that the number of interruptions experienced by Ontario Hydro’s “average” customer would remain
unchanged. For the distribution system, a similar approach is used. Historical values both for energy
unsupplied as a fraction of peak demand and for the fraction of interruptions of a given duration are
assumed to be constant into the future. Thus, the “average” customer would continue to experience one
or two interruptions per year.
The costs of unreliability are considered to be of two types: direct costs to customers and indirect costs
incurred by other elements of society. The direct costs incurred by each customer are affected by the
frequency of outages, the duration of each outage, the magnitude of each outage, the amount of ad-
vance notice, and the time of the day, week, or year when each outage occurs. However, the reliability
indices used by Ontario Hydro are the mean (or average) values of frequency, duration, and unsupplied
energy. Thus, of the factors that affect customer costs, only the mean values of the first two (frequency
and duration) are computed as reliability indices. The mean value of energy not supplied may be
determined from the frequency, duration, and magnitude of generation deficiencies. To estimate the
direct costs of unreliability, Ontario Hydro’s customers are divided into four groups: large users, other
manufacturing, commercial, and residential and farm. These groups are associated with the existing
rate structure. The direct (or out-of-pocket) costs for the customers in the first, second, and fourth
groups are based on Ontario Hydro surveys asking for estimates of direct losses as a result of interrup-
tions of different durations that occur without warning. The questions asked required a considerable
effort on the consumer’s part to provide reasonable numerical answers. So, clearly, the results should be
taken as indicators rather than as highly precise values. Nevertheless, they give some insight into the
magnitude of the cost of an outage in relation to its duration. No survey was available for the commer-
cial group, and so its costs were estimated on the basis that the losses are proportional to wages and
salaries in the commercial sector of Ontario’s economy and to the amount of energy not supplied on
demand. It was also assumed that the costs of interruptions for the farm group are the same as those for
the residential group. The results of the surveys are summarized in Figure 4.4.
The next step isto combine the loss data for each customer class, to make an estimate of the total cost of a
given level of unreliability. First, the losses caused by the generating system are estimated. This is
done by assuming that all load-shedding is imposed on customers one hour at a time and that there is no
discrimination among customers, that is, that each customer group experiences an equal number of
one-hour load cuts. The direct losses caused by generation unreliability are calculated by combining the
four customer loss functions of Figure 4.4 with the estimates of load-shedding produced by the F&D
computation.
It is important to note that for a given planned reliability level, the uncertainty in load forecast and
generation in-service dates means a considerable variation in reliability level that may be achieved ina
given future year. Since the cost of outages escalates quickly as the shortfalls in generation increase,
the costs of outage computed as if the planned reliability level had been achieved are not adequate
indicators of the real impact. Thus, for a given planned reliability level, it is the expected value of the
cost of outages (that is, the mean of the costs over all probable reliability levels) that is computed, not the
cost corresponding tothe expected level of reliability.
Customers’ direct losses due to transmission system unreliability were computed by assuming that
each customer class had equal exposure to interruptions and that each would experience its share of
The Reliability of the Electric Power System 51
long and short interruptions. The customer losses attributable to the distribution system are computed
by assuming that all have equal exposure to interruptions (except those connected directly to the
transmission system).
Table 4.3 shows the direct losses by customer class attributable to the generation, transmission, and
distribution subsystems. Note that, among the classes of customers, "other manufacturing” has the
largest share of total losses. Moreover, distribution accounts for more direct costs of interruption than
transmission. Table 4.3 also illustrates the contrast between the share of Ontario’s electricity that is
consumed by each customer class and the computed share of total direct losses accruing to each class.
Table 4.3 Projected Direct Customer Losses in 1990 — SEPR Study?
Average direct customer losses
(millions of 1990 dollars) Share of total
Generation Transmission Distribution Total direct losses (%)
Share of electricity 23% 28% 23% 28% 23% 28%
Customer class demand (%) reserve reserve reserve reserve reserve reserve
Large users 31 125 c 3.2 2.50 18.2 | 28 18
Other manufacturing 19 12.5 ¢ 2.9 13.1 28.5 16 43 52
Commercial 17 93 ¢ 1.6 73 18.2 8.9 28 29
Residential and farm 33 0.3 c 0.1 ‘92 0.6 0.3 1 1
Total 100 34.6 0.08 78 23.1 65.5 30.9 100 100
Notes:
a) Projections are for a 5.5 per cent average annual rate of load growth to the year 2000.
b) Many large users are supplied directly from the bulk power transmission system.
c) Relatively small.
Sources: RCEPP and Ontario Hydro, “System Expansion Program Reassessment Study”’, Fifth Interim Report, November 1978.
The indirect costs of interruptions are taken to be the reduction in the gross provincial expenditure
(GPE), over and above the direct costs in the industrial and commercial sectors. The direct losses
imposed on these customers are assumed to cause a reduction in profits, and 60 per cent of this reduction
in profits is taken as the reduced investment by these two sectors. Through an econometric model, the
reduction in investment is translated into a reduction in the GPE. The difference between these two is
taken as the indirect cost of interruptions to supply. The total economic cost of a given level of reliability
is then the sum of the direct costs incurred by the residential and farm customers, 40 per cent of the
direct costs incurred by the industrial and commercial customers, and the reduction in GPE that is
associated with the reduced investment by the industrial and commercial sectors.
As shown in Figure 4.5, the total economic costs attributed to a given planned reserve margin are the
sum of the present value (over the study pericd) of Ontario Hydro’s expenditures for system expansion
and the total economic costs of interruptions. Figure 4.5 shows that the effective target generation
reserve corresponding to minimum total economic cost considerations is about 5 per cent lower than the
reserve determined by the traditional LOLP criterion of 1/2,400. The figure also shows that the cost of
interruptions drops rapidly as the reserve margin increases and is virtually nonexistent at the “‘opti-
mum” reserve level. The choice of “optimum” reserve is influenced by many factors which have not
been included in the computations of the SEPR study. Therefore, results such as those presented in
Figure 4.5 should be interpreted as indicative rather than definitive.
The studies that try to balance the costs to the customer with the costs to Ontario Hydro are an impor-
tant improvement in the effort to determine a justifiable level of system reliability. However, there are
several aspects of the SEPR approach that require further investigation. First, the customer surveys
ask for a single estimate for each category of cost and duration. These estimates are inherently uncer-
tain, and it is suggested that the respondents should be allowed to respond with a range of costs. Then
some estimate of the range of uncertainties might be given on a graph similar to Figure 4.5. Second, to
develop scenarios of generation expansion, the study assumes a reserve margin whose value would be
constant over the planning period. A constant reserve margin has not been the planning criterion in
the last few years; a better measure, the LOLP, has been used. Moreover, the system’s planned reserve
margin has not been constant in the past. With the proportion of hydraulic and nuclear plants in the
system due to change significantly in the future, the reserve margin should not be held constant. We
feel that a balance of customer and utility costs on a smoothed annual basis, and not a constant reserve
margin with the lowest present value of costs over the whole study period, should be a planning objec-
tive. For example, as shown in Figure 4.6, the annual energy not supplied at reserve margins of 17 per
52 The Electric Power System
Fig. 4.5: p. 56
Fig. 4.6: p. 57
cent and 22 per cent is growing much faster than the system peak. Thus, a present value calculation
assuming constant reserve margin may conceal the fact that the costs of interruption are escalating
quickly, and it may not indicate that the reserve margin should perhaps be increased as time pro-
gresses, rather than being held constant. Third, the extra costs incurred by the utility to increase the
reserve margin from the base case are computed by adding 800 MW coal-fired units. Clearly, extra
capacity can also be obtained by advancing the nuclear programme (especially in the case of low nuclear
scenarios) and/or by adding some combustion turbine units or smaller coal-fired units. Since the slope
of the “expenditures” line in Figure 4.5 significantly affects the balance point, the effects of using other
types of plants to increase capacity should be explored. Finally, as noted in the SEPR study, further
work needs to be done to investigate transmission and distribution reliability in more detail.
Summary and Conclusions
The assessment of the reliability of an electric power system is a very complex and difficult problem.
Ontario Hydro, like many North American utilities, has traditionally used the probability-based LOLP
approach for evaluating the reliability of its generation plans, and security criteria for planning bulk
power transmission. The planned generation reserve margins have been in the order of 30 per cent to
correspond to an LOLP of 1/2,400. Due to the inability of the LOLP method to provide measures of
reliability in terms of the frequency, duration, and magnitude of outages, Ontario Hydro has developed
a generation reliability programme based on the frequency-and-duration-of-outages method. The
F&D technique is considered to be far superior to the LOLP method.
While methods of evaluating generation reliability have received most of the attention of utility plan-
ners, Ontario Hydro and its neighbours are active in developing practical schemes for the quantitative
evaluation of transmission reliability. These schemes are necessarily complex, and it will be some time
before they find widespread use among electricity utilities.
In Ontario, the contribution of failures in the bulk power system to interruptions of supply to customers
has been small. Most interruptions occur as a result of failures in the distribution system. For example,
data from Ontario Hydro’s rural customers show that they suffered interruptions with a total average
duration of 182 minutes per year, whereas the duration of supply interruptions at each bulk power
delivery point was only 21 minutes per year.
Ontario Hydro has undertaken studies to assess reliability by balancing its costs and benefits to the
consumer. A measure of the benefits of reliability, which are extremely difficult to estimate, was ob-
tained by canvassing customers to get their perception of losses due to supply interruptions. Further
work needs to be done, but the results of these studies already indicate that the planned generation
reserve margins could be reduced by a few percentage points. As a result of its studies, Ontario Hydro
has reduced its planning reserve from approximately 30 per cent to about 25 per cent. The new target
reserve is based on a reduced generation reliability criterion of 10 system minutes of unsupplied
energy per year, after taking due account of a 2.7 per cent load reduction by voltage reductions, and of
some help (500-700 MW) from interconnections.
Studies such as these constitute an important improvement in the efforts to determine a justifiable level
of system reliability. They indicate that Ontario Hydro is among the pioneers in advancing the state of
the art of reliability assessment. It must be noted, however, that Ontario Hydro has not attempted to
evaluate, from a reliability viewpoint, decentralized system scenarios based on the concept of “‘local
energy centres”. A factor that may be hindering this attempt is the unavailability of practical techni-
ques for evaluating overall system reliability.
The Reliability of the Electric Power System 58
Figure 4.1 Variation in Loss of Load Probability with a Change in Reserve Margin (Ontario Hydro East System)
10/2,400
Loss of load probability
Moin taal Uo oe Paka
Reserve shortage Reserve surplus
Variation in percentage reserve margin
Note: Based on data from a proposed 20-year expansion of Ontario Hydro’s East System under long-range forecasts LRF 48 and LRF 48A. The reference value for reserve variation
corresponds to an LOLP of 1/2,400.
Sources: RCEPP and Ontario Hydro.
Figure 4.2 Variation in Incremental Reserve Margin with a Change in the Forced Outage Rate and Unit Size
A—Unit size = 750 MW
B—FOR = 12% A B
1,250
12
10 1,000
Forced outage rate (%) Unit size (MW)
8
750
6:
10 20 30 40
Reserve as a percentage of additional load
Source: “Generation Planning Processes”, Ontario Hydro submission to RCEPP, May 1976, Exhibit 21.
54 The Electric Power System
Figure 4.3 Possible Reduction in Reserve Requirements with Interconnections
6,000
5,000
LOLP = 1/2,400
4,000
Reserve requirements of
each system (MW) LOLP = 1/240
3,000
LOLP = 1/24
2,000
1,000
0 ;
0 1,000 2,000 3,000 4,000 9,000
Interconnection capacity (MW)
Source: “Reliability”, Ontario Hydro submission to RCEPP, May 1976, Exhibit 20.
Figure 4.4 Customer Loss Functions ' — SEPR Study
50
Other manufacturing
40
Notes:
1. Customer loss functions relate customers’ direct financial losses
caused by interruptions that occur without warning,
to the duration of the interruptions.
2. The residential customer loss function rises from 0.03 dollars per
30°- kilowatt for momentary interruptions to 0.12 dollars per kilowatt
for 16-hour interruptions.
Direct customer loss in 1976
dollars per kilowatt of load
interrupted
20°
Commercial
Large users
10 =
0 Residential?
SG ae Le SE a eT a
0 4 8 12 16
Duration of interruption (hours)
Source: “System Expansion Program Reassessment Study”, Ontario Hydro, Fifth Interim Report, November 1978.
The Reliability of the Electric Power System 55
Figure. 4.5 Variation of Total Economic Costs with the Target Generation Reserve — SEPR Study
4%
Total economic cost
2%
0%
LOLP = 1/2,400
—2%
Expenditures
Total economic costs as a percentage of costs at the current reliability standard
—4%
°
Costs of direct and indirect losses
20% 20% 30% 39%
Effective target generation reserves on the East System
Note: Based on present value of costs for 1985-1997 at a 15 per cent discount rate and a 5.5 per cent annual growth in load.
Source: “System Expansion Program Reassessment Study”, Ontario Hydro, Fifth Interim Report, November 1978.
56 The Electric Power System
L661
VOIINGIAS|q
UOISSIWSUe!|
anasal %9¢
anlasal YEE
aAlasal %97
anlasal %¢2
“SUIBJEW BAJBSAJ J9BJ2} SNOLEA Je (ABJauUa paljddnsun je}0} pue) uoIJesUa5
aNlasal %Q] —_*
19,
C661
8/6] J8qWanoy ‘WOday WHa}u] YY!4 “OIPAY OLeIUQ ‘APNG JUaWSSasseay WeIdO1g UO|SUedX WaySKs,, :aounos
‘YeaK Jad U9d Jad ¢°G JO 8}21 YJMOUB peo] e je BuIpUedXa WA}SAS e JO) ale SaJeUI!]S3 :a}ON
L861
001
002
(Y © MW $0 Spuesnouy) A}ljiqeijauun UOJeJQUed pue ‘UOISSIWSUA. ‘UOIINGLISIP Jo asnedeq paijddns Jou AgJaua Jo JUNOWe jenuUe aBeIaAy
00€
APMIS YdIS — Wa}SAS S23 OIPAY OLE}UG BY} UO AYijiqeI|a1UA] WA}SKS jelo] g'p “aunBLy
The Reliability of the Electric Power System 57
a —— a, = (a =
pans ghitedes ea at .-
ge : va : = >
es “ -
a rae J a - a
CHAPTER FIVE
Interconnections with Other Systems
Ontario Hydro’s province-wide power transmission system is interconnected with those of electricity
utility systems in neighbouring provinces and states, and these systems are, in turn, similarly inter-
connected with their neighbours.
The gradual evolution from small isolated systems in the early years of the industry to today’s interde-
pendent systems reflects the growing recognition that reliability of service can be improved and cost of
service reduced through interconnection and co-ordination.
Co-ordination of the planning and operation of electricity supplies on interconnected systems offers a
variety of benefits. Hydroelectric resources may be utilized more fully if developed to serve regional
rather than local needs. Nuclear or fossil-fuelled generating stations may be located at advantageous
sites, and concentrated in large centres so as to realize economies of scale. The amount of generating
capacity needed by each interconnected system may be reduced, because the peak demands of the
participating utilities vary in size and timing — the so-called load diversity (see Figure 2.1). Require-
ments for reserve generation may be reduced by special arrangements for mutual assistance in emer-
gencies. Responsiveness to unanticipated shortfalls or overruns of supply and demand is improved. The
reliability of the bulk power system may be enhanced and service standards improved by the co-
ordination of spinning reserves, system frequency and voltage controls, operating schedules, and
maintenance programmes. Substantial reductions in operating costs may be realized through co-ordi-
nated operation of all the production facilities available to the interconnected systems. The capability
for dealing with extreme contingencies and disasters is improved. Financial burdens may be alle-
viated through the sharing of risks as well as of benefits. The move towards fuller utilization of renew-
able primary energy resources may be accelerated, and dependence on scarce and dwindling supplies
reduced.
On the other hand, the establishment of an interconnection can be costly, particularly if it is of substan-
tial length. There must be assurance of interconnection transactions of large and sustained value to
warrant the investment. Other costs may be incurred, including those resulting from the environmen-
tal impact of the interconnection and of the related power plant construction and operation. And the
various participants may find it difficult to reach agreement on the benefits and costs and how they
should be allocated among the interests affected.
The use of interconnections also raises complex issues in respect of public policy. Consider, for example,
the transmission lines across the Niagara River that interconnect the Ontario Hydro and Niagara
Mohawk systems. The power that has been transferred over these lines has provided substantial eco-
nomic benefits to both parties. However, the existence of this interconnection was a major factor
contributing to the massive black-out of 1965, in which almost the entire Atlantic seaboard was
plunged into darkness. That the utilization of such interconnections raises important questions of
public policy is evidenced by the fact that transactions concerning them may require the approval, not
only of the utilities concerned, but also of the provincial, state, and federal governments and their
agencies.
Electricity Trade with Neighbouring Provinces and States
Existing Interconnections
Ontario Hydro has high-voltage interconnections with electricity utilities in neighbouring provinces
andstates, as follows: |
Manitoba Hydro — one 115 kV and two 230 kV lines near Kenora, providing a nominal interchange
capability of 475 megavolt amperes(MVA).
Hydro-Québec — eight 115 kV and four 230 kV lines that cross the Ontario-Quebec border at various
points (Beauharnois, Masson, Ottawa, Chats Falls, New Liskeard, Kirkland Lake), providing a nomi-
nalinterchange capability of 2,735 MVA.
Detroit Edison Co. — two 230 kV and two 345 kV lines at Sarnia and Windsor, providing a nominal
interchange capability of 2,855 MVA.
Interconnections with Other Systems 59
Power Authority of the State of New York (PASNY) — two 230 kV lines at Cornwall, providing a nominal
interchange capability of 720 MVA.
Niagara Mohawk Power Corporation and PASNY — two 230 kV lines at Niagara, providing a nominal
interchange capability of 720 MVA.
The actual transfer capability depends upon overall configurations of load, and generation and trans-
mission capacity in the vicinity of the interconnection, and is usually well below the nominal capacity. It
also varies with the direction of power flow. Ontario Hydro also has interconnections with other power
systems within Ontario, which will be discussed later, and it deals, as circumstances warrant, with other
utilities in the U.S. with which it is not directly connected, by arranging with the intervening utility
systems to “wheel” the power. Through these arrangements, Hydro has access to a large and dynamic
wholesale market.
The Volume of Transactions
Table 5.1 shows Ontario’s annual electricity trade with its neighbours (1968-77) in billions of kilowatt
hours. These are totals for the province but comprise mainly transactions made by Ontario Hydro.
Electricity trade was vigorous throughout the period. In recent years, total trade (exports plus imports)
has approached 25 per cent of the total electricity made available in Ontario (production in Ontario plus
imports less exports). Imports exceeded exports and came mainly from postin exports went mainly to
the United States.
Table 5.1 Ontario — Interprovincial and International Electricity Transfers
Wefurmee of Grade (TWN) 2 os Ne gt ee ee ee ee
Quebec
Imported by Ontario 48 47 6.3 5.9 8.1 9.6 10.8 11.4 i? 10.7
Exported by Ontario 0.2 0.1 0.2 0.3 0.2 0.1 0.2 0.3 0.3 0.3
Manitoba
Imported by Ontario 0.1 0 0.2 0.3 0.5 0.8 1.6 17 1.6 0.9
Exported by Ontario 0.1 0 0.8 0 0 0 0 0 0.1 0.1
U.S.A.
Imported by Ontario 2.6 22 2.9 2.6 Ne 1.6 1.8 ae, zt 1.2
Exported by Ontario Zo 2.9 3.6 4.1 6.1 7.6 19 48 6.2 9.6
Totals
Imported by Ontario is 6.9 9.4 8.8 10.3 12.0 14.2 15.8 15.4 12.8
Exported by Ontario 2.8 3.0 4.6 44 6.3 ay 8.1 | 6.6 10.0
Total imports and exports 10.3 9.9 14.0 13.2 16.6 19.7 vee 20.9 22.0 22.8
Total available in the province
(production plus imports-exports) 61.1 64.8 69.5 73.0 79.1 84.3 88.8 89.2 96.1 98.6
Percentage of total available
Imported by Ontario 12 11 14 12 13 14 16 18 16 13
Exported by Ontario 5 5 7 6 8 9 9 6 7 10
Total imports and exports 17 16 21 18 21 23 25 24 23 23
Cost and revenue
hl ata ise Rs Se ae eT TAO SEE ME BS oR DI) Sl Ie ie ret ek a Sal tie = ee ol ete eh atte Slee de Dt ele ite Pi
Imports by Ontario
Cost of imports ($million) 15.1 14.3 28.6 29.4 25.7 39.5 57.3 64.7 66.5 65.1
Unit cost (cents/kW:h) 0.20 0.21 0.30 0.33 0.25 0.33 0.40 0.41 0.43 0.51
Exports by Ontario
Revenues from exports ($million) 44 9.6 22.9 28.4 41.0 72.4 105.2 48.0 94.9 216.5
Unit price (cents/kW-h) 0.16 0.32 0.50 0.65 0.65 0.94 1.30 0.94 1.44 214
Source: “Total Transfers by Ontario Utilities and Industrial Establishments’’, Statistics Canada 51-202, vol. 2.
Table 5.1 also shows the cost of imports and the revenues from exports. The costs of imports exceeded
the revenues from exports in the late 1960s and early 1970s, but the situation has been reversed since
then. The unit price of exports has, in fact, always been considerably higher than the unit cost of
imports, except in 1968. While the unit cost of imports escalated at a relatively moderate rate from 0.2 to
0.5 cents/k W-h, the unit price of exports soared from 0.2 to 2.2 cents/k Wh.
Ontario Hydro’s imports have been mainly in the form of purchases of firm hydroelectric energy from
Quebec, and to a lesser extent from Manitoba, on a long-term basis and at a fixed price that is low by
present standards. Its exports have been mainly through short-term opportunity sales of non-firm
energy from coal-fired generation that is temporarily surplus to its own needs, to American utilities
a a
60 The Electric Power System
that would otherwise have to meet their requirements from higher-priced oil-fired generation. The
deliveries of firm hydroelectric energy from Quebec ended in 1977, however, and Ontario Hydro’s
transactions with neighbouring states and provinces are now mainly for the purchase and sale of non-
firm, or ‘‘secondary”, energy.
Types of Interconnection Transactions
Ontario Hydro has interconnection agreements that prescribe the ground rules under which it ar-
ranges short-term transactions with its neighbours. These transactions are of the following types:
Short-Term Capacity. Power and associated energy provided for a period of from one week to about six
months to supplement the peak generating capability of the receiving party during periods of pro-
longed capacity deficiency.
Daily Capacity. Power and associated energy provided on a day-to-day basis to supplement the peak
generating capability of the receiving party at atime of temporary capacity deficiency.
Spinning Reserve. Excess generating capacity that is held ready by one party for use by another on a
few minutes’ notice.
Supplemental Energy. Energy provided during off-peak hours to supplement storages (fossil-fuel or
hydraulic) of the receiving party.
Economy Energy. Energy delivered in order to effect a saving in the cost of generation when the
receiving party has adequate generating capability available to carry its own load.
Tertiary Energy. Scheduled deliveries of energy that cannot be classified under other categories.
Inadvertent Energy. Non-scheduled energy.
Wheeling Service. The transportation of electric energy on ascheduled basis through one party’s system
at the request of the other party.
Price schedules for these transactions are reviewed periodically. Economy energy is usually priced ona
split-the-savings basis, that is, mid way between the supplier’s incremental cost of fuel, incidental
materials, and labour and the buyer’s ‘“‘decremental cost”, or savings, in reduced fuel and incidental
costs.
A typical capacity charge is $100 per megawatt day, subject to periodic revision. The energy associated
with a capacity sale may be sold at the economy energy rate or at cost plus 10 per cent, whichever is
higher. It is advantageous to the seller to collect both a capacity charge and an energy charge, and most
of Ontario Hydro’s sales have been of this nature.
The Outlook for Electricity Trade
Like other aspects of the energy sector in general and the electricity supply industry in particular,
interconnection practices are today receiving close scrutiny from numerous diverse interests, and the
course of future development is in doubt.
Most transactions with U.S. utilities have been short-term opportunity sales from coal-fired generation
that has been temporarily surplus to domestic needs. Sales have been large and highly profitable. The
selling price has been substantially more than either the incremental cost of production or the whole-
sale rate to domestic customers.
Ontario Hydro, at present, imports most of the coal it needs from the United States. A portion of these
imports is used for the power exports to the U.S., and some experts maintain that this two-way trade
helps to ensure the availability of coal from the U.S. Others object to these transactions on the grounds
that they aggravate atmospheric pollution in Ontario. However, in approving such exports, the Na-
tional Energy Board has expressed satisfaction that the revenues adequately cover ail costs incurred,
including social costs.
Ontario Hydro is faced with considerable surplus in generating capacity, because demand has not
grown as much as expected, and this surplus is expected to continue until the early 1990s. The surplus
power will peak at about 4,000 MW in the early 1980s. Hydro is striving to make profitable sales from
this surplus and has had some success — six-month contracts ending in December 1979 were signed
with the Niagara Mohawk Power Corporation (250 MW) and General Public Utilities (200 MW).!
It is obviously not possible to forecast such opportunity sales beyond the short term with any confidence.
Snr nnn nnn UU ttdtttIdtIddtSdSIIIS IS SSSSSSSSSS
Interconnections with Other Systems 61
Ontario Hydro is commencing to use higher-cost western Canadian coal to reduce its dependence on
United States coal. The price of exported energy will increase to the extent that western Canadian coal
is charged to the transaction, and this may have a dampening effect on demand. U.S. utilities’ expecta-
tions of growth, like Ontario Hydro’s, have declined, and this tends to reduce sales opportunities.
Conversely, the sheer size of the U.S. market, its heavy reliance on oil-fired generation, and the diffi-
culties U.S. utilities are encountering in commissioning coal-fired and nuclear plants all suggest a
continuing buoyant market for Ontario’s surpluses. Hydro has indicated that the most promising
markets for firm power sales appear to be the Central Area Power Co-ordination Group, the Michigan
Electric Power Pool, the American Electric Power Service Corporation, and some members of the New
York Power Pool, with a total potential in the 500 M W to 1,500 MW rangein the 1980s.”
Superficially, there is no good reason why the energy in U.S. coal should continue to flow into Ontario for
processing and then return to the U.S. in the form of electricity. One might expect that the U.S. could
produce electricity from its own coal more economically at home. Institutional biases encourage such
flows across the border, however.
The U.S. electricity utility industry is largely investor-owned, profit-oriented, and subject to taxation
on income and to constraints on capital funding. While the industry is much larger than Canada’s in
aggregate, it is more fragmented. Some 200 investor-owned utilities own about 75 per cent of the
generating capacity, five large federal agencies have 12 per cent of it, and the remaining 13 per cent is
owned by hundreds of small investor-owned utilities and by state, municipal, and other non-federal
agencies.
The Ontario Hydro system is larger and more diversified than any of the U.S. utilities with whom it
deals. Its plant is more capital-intensive and has relatively lower incremental production costs. The
system comprises mainly hydraulic, nuclear, and large and efficient coal-fired steam turbine genera-
tion. Its U.S. neighbours, on the other hand, have a large share of plants with high fuel costs, such as oil-
fired, gas-fired, and less efficient coal-fired plants. So, for economic reasons, there is a tendency for
power to flow from north to south. Export opportunities may be expected to remain at a high level, at
least until the U.S. utilities substantially increase the mix of their nuclear and coal-fired capacity.
Similar considerations apply all along the Canada-U.S. border, and over the last few years a number of
provinces have stepped up their power exports to the United States. British Columbia, Saskatchewan,
Manitoba, Quebec, and New Brunswick now have substantial interchange capability with contiguous
American utilities. There is an understandable preference to market southwards into the higher-
priced U.S. market, particularly since the requisite interconnections are shorter and less costly than
those between provinces.
The era of long-term fixed-price imports of firm hydroelectric power from neighbouring provinces
appears to have ended, at least for a few years. The last such contract with Quebec terminated in 1977.
Ontario’s only remaining major firm-power contract provides for the purchase of about 200 MW from
Manitoba Hydro each year up to and including 1981. Ontario Hydro now plans well ahead, to ensure
that it will have adequate generating capacity for its needs. Indeed, with the prospect of a substantial
surplus for most of the coming decade, Ontario is unlikely to buy anything but secondary hydroelectric
energy from Manitoba and Quebec during that period.
On the face of it, it appears desirable for Ontario Hydro to resume the purchase of firm hydroelectric
power from its neighbouring provinces on a medium-term basis, as soon as it is in a position to do so.
Substantial hydroelectric resources remain to be developed in Quebec, Labrador, and Manitoba, it
appears desirable, especially in the interests of conserving non-renewable resources, for them to accel-
erate the development of these resources and use the resulting capacity to supply other markets,
including Ontario, until this capacity is required for their own purposes.
However, hydroelectric projects in the remote regions of Canada tend to be highly capital-intensive,
which may present problems in financing them. Also, the unit capital cost of an additional station may
be less than the average unit capital cost of the whole project. In these circumstances, it is difficult to
price medium-term sales of firm power on a mutually acceptable basis. Furthermore, Manitoba and
Quebec may prefer to sell to U.S. utilities rather than to Ontario Hydro, if they can obtain a better price
from them. This preference will be reinforced, if, as is generally the case, the interprovincial intercon-
nections are more costly than the international ones. However, the total economic benefit to Canada
may be increased, and broader regional or national goals may be better served, if the interprovincial
62 The Electric Power System
transaction proceeds. Just as with other major transportation and communication links, there is an
element of nation-building in the establishment of interprovincial interconnections.
Another uncertain factor that will affect future interconnection transactions is the availability of
power transfer capability. Ontario Hydro plans to add a 345 kV interconnection at Niagara in 1983.
Hydro estimates that this will increase its total transfer capability to the United States to about 3,000
MW.? At the Commission’s hearings on bulk power facilities for southwestern Ontario, it noted that the
transfer capability will decline after 1983 unless new bulk power transmission facilities are installed
in southwestern Ontario. The existing main transmission lines westward into London that supply loads
in the London-Windsor-Sarnia area also transmit some of the power to be exported to Michigan. To
avoid overloading these circuits, the export capability drops as the domestic loads increase.
In similar vein, at the Commission’s hearings on bulk power facilities in eastern Ontario, Ontario
Hydro showed that the growing load in the Ottawa area will restrict the amount of power that can be
transferred tothe New York Power Pool at Cornwall, starting about 1988.4
In order to reinforce the transfer capability into the United States, it appears that it may be necessary
to strengthen Ontario Hydro’s main transmission into southwestern Ontario or into eastern Ontario,
or both. Other possibilities may include a submarine cable under Lake Erie from the Nanticoke Gener-
ating Station, an interconnection via the Great Lakes Power Company at Sault Ste. Marie, or a further
strengthening of the Niagara Falls crossing. Any such step would, of course, depend upon system
arrangements on both sides of the border.
The Beauharnois tie from Quebec can still perform a valuable role in the transport of any available
power from Quebec. However, it is of limited utility for transferring power from Ontario, since its use
for that purpose would jeopardize security of supply to Ottawa.
Because the Ontario-Quebec ties are, for all practical purposes, only a one-way interconnection (Quebec
to Ontario), Ontario Hydro considers that valuable opportunities for economic interchange and mutual
support between the two systems are not being exploited fully. There has been considerable discussion
between Ontario Hydro and Hydro-Québec concerning the construction of a high-voltage, direct-
current link between thetwosystems, but, before such an interconnection could materialize, considera-
ble strengthening of the Ontario Hydro system in eastern Ontario would be required. Hydro-Québec
may prefer to establish such a link with the Power Authority of the State of New York instead of with
Ontario Hydro, and it is even conceivable that some type of tripartite configuration might be devised.
Transactions with Manitoba are necessarily on a relatively small scale that can be accommodated by
Ontario Hydro’s West System; Hydro’s internal ties are too weak to permit the participation of its much
larger East System. A strong transmission link will be required between the West System and the East
System to permit the generating capacity in these two systems to be more fully co-ordinated, and to
permit increased transfers with the western provinces.
The Role of the National Energy Board (NEB)
The export of electricity from Canada requires a licence from the National Knergy Board. At one time,
Canada was reluctant to authorize exports, and they were subject to duty from 1925 until 1963. Since
1963, Canada has viewed electricity exports more favourably, and the government is prepared, upon
recommendation by the NEB, to authorize exports when a neighbouring utility needs emergency
assistance or when there is an opportunity to sell surplus power profitably. An applicant for an export
licence must demonstrate to the NEB that the proposed electricity sales are surplus to foreseeable
Canadian requirements that could reasonably be supplied by the applicant and that the price to be
charged is just and reasonable in relation to the Canadian public interest. The applicant must also
provide NEB with details of any environmental impact that may result from the generation of the
power for export.
In the last few years, there has been increased participation in NEB hearings concerning interna-
tional power lines and electricity exports. Neighbouring provinces and utilities, property owners, en-
vironmentalists, and ratepayer groups have intervened, some in support of and some in opposition to
the applicants. Neighbouring provinces have not opposed export applications, but on occasion have
sought to secure the right of recapture for their own use in certain eventualities. Licences issued by the
NEB generally stipulate that the export of interruptible energy be stopped or curtailed whenever and
to whatever extent such energy is required to supply any firm load in Canada, or if any Canadian
electricity utility is willing to buy part or all of the energy at the same price as that of the export,
RD
Interconnections with Other Systems 63
adjusted for possible differences in the cost of delivery. However, the NEB has expressed concern that it
may be placed in an invidious position in adjudicating between two provincial utilities, on the question
of Canadian interest, if and when one utility becomes deficient in generation through lack of adequate
planning and through no fault of the other.
“Blectricity Exchanges” — A Canada-United States Study
In 1979, the governments of Canada and the United States jointly issued “Electricity Exchanges”, a
report on a study that examines the potential for increasing electricity exchanges between the two
countries.
The participants in the study were regional representatives of the electricity utilities along the United
States-Canada border, federal officials of both countries, and representatives of the Canadian provin-
cial governments except for Quebec, which declined an invitation to participate. Observers from the
Quebec government and Hydro-Québec participated in organizational meetings.
Representatives of the Ontario Ministry of Energy and Ontario Hydro participated in the study group
for the Ontario-New York-ECAR region. ECAR is the East Central Area Reliability Co-ordination
Agreement, one of the regional reliability councils that make up the U.S. National Electric Reliability
Council. It includes utilities in eight states, including Michigan, Ohio, and Pennsylvania.
For the Ontario-New York-ECAR region, the report on the study notes that the expected transfer
capability in 1980 is as shown in Table 5.2. But the report also states that these transfer capabilities will
decline unless the Niagara interconnections are strengthened and the internal transmission capacity
constraints within the Ontario-to-Michigan interface are overcome. The report concludes that if these
transmission limitations are resolved, there is considerable potential for increased operational co-
ordination to stimulate economic exchange — for example, the use of excess coal-fired generation in
Ontario to reduce oil consumption in New York and Michigan and in utilities farther south. In addition,
there may be significant possibilities for seasonal diversity exchanges between Ontario and utilities to
the south of New York and Michigan.
Table 5.2 Expected Power Transfer Capability between Ontario and the U.S. in 1980
Summer transfers Winter transfers
Transfers to U.S. (MW) to Ontario (MW)
Ontario and Michigan (zero to New York) 1,200-2,000 600—1,100
Ontario and New York (zero to Michigan) 1,500—2,200 1,300—2,000
Ontario, Michigan and New York (simultaneous) 2,000—2,700 1,700-2,400
Note: Transfer capabilities shown are based on a range of 0—400 MW circulating current around Lake Erie in a counter-clockwise direction.
Source: ““Canada/United States: Electricity Exchanges’’, U.S. Department of Energy and Energy, Mines and Resources Canada, May 1976.
In order to realize fully the benefits of interconnections, “Electricity Exchanges” suggests, utilities
near the U.S.-Canada border should maximize co-ordination of system planning and operation, expand
the sharing of technical information, ensure that applications for electricity exports are filed in a
timely manner, and develop mutually agreeable interchange rates and wheeling rates in the U.S. to
encourage participation by utilities not directly adjacent to the border in international electricity
trades. The report also recommends that the federal, state, or provincial governments and regulatory
agencies should clarify government policies relating to firm exports and exports in general, effect
increased communications among regulatory agencies and electricity utilities to expedite regulatory
approval, develop public information programmes, and ensure that pricing policies are consistent with
a fair sharing of interconnection benefits.
Interprovincial Interconnections
The Provincial Utility Systems
Because of the provisions of the British North America Act, which gives each province jurisdiction over
the natural resources within its boundaries, the production and distribution of electricity is subject
mainly to the provincial jurisdictions. Each province has developed a province-wide transmission sys-
tem that interconnects most of its generation and load centres. These province-wide systems permit
generation to be planned, constructed, and operated in a co-ordinated manner, so as to meet provincial
requirements for electricity as economically as possible.
64 The Electric Power System
Canada’s immense distances and thinly distributed population make the forging of interprovincial
links acostly undertaking. However, numerous important links have already been put into place, anda
number of additional major links are receiving active consideration through bilateral discussions
between the utilities directly concerned.
The Federal Government
The federal government has a large and growing presence in electricity matters. It regulates electric-
ity exports through the NEB. Through the Northern Canada Power Commission, it produces and
markets electricity in the Yukon and Northwest Territories; and through Atomic Energy of Canada
Ltd. it owns several nuclear power plants and has a financial interest in others. It has granted funds
towards the construction of some provincial generation and transmission facilities and has invested
funds in some others. Federal statutes and policies influence provincial electricity utility affairs in
numerous other ways. For example, the federal government is responsible for international boundary
waters and navigable waters and for the regulation and control of some aspects of fossil and nuclear
fuel production, transportation, and utilization.
The federal government probably has the constitutional authority to regulate interprovincial electric-
ity trade if it chooses to do so. There appears to be little likelihood, however, that it would take any
unilateral action in this respect. It has preferred to foster strengthened interprovincial interconnec-
tions in close co-operation with the provinces concerned.
Federal-Provincial Studies
From time to time, the provincial and federal governments have given preliminary consideration to
the establishment of high-capacity interprovincial interconnections that would result in a Canada-
wide power network.
A federal-provincial working committee on long-distance transmission made studies during the pe-
riod 1962-7. While these studies suggested that the interconnection and co-ordinated development of
the power systems of all the provinces would be economically beneficial, it was concluded that such
benefits were marginal. Further action was deferred.
The energy crisis of 1973 renewed interest in the idea of a Canada-wide power grid. While priority
consideration has necessarily been given to oil, the federal and provincial energy ministries have seen
strengthened electricity interconnections as a potential part of a balanced policy for promoting the
conservation and prudent utilization of all forms of energy. The federal government recognized the
importance of strengthening interprovincial interconnections in a 1976 report — ‘An Energy Strat-
egy for Canada: Policies for Self Reliance” — and advocated the acceleration of such developments and
closer co-ordination in the joint planning and development of power projects by the provincial utili-
ties.° The report also noted:
In January of 1974, the Minister of Energy, Mines and Resources announced that the Government of
Canada would pay 50 per cent of the cost of approved studies relating to interprovincial or interre-
gionalelectrical interconnections and finance up to 50 per cent of the capital cost of approved projects.
Thus, it is reasonable to assume that any initiative by the provinces to strengthen interprovincial inter-
connections is likely to enjoy the support of the federal government.
Interprovincial Advisory Council on Energy (IPACE)
During 1978, the Interprovincial Advisory Council on Energy, which brings the provincial deputy
ministers of energy together in conference, sponsored a preliminary study comparing the benefits to
be gained from continued normal inter-utility development with those that could be gained through
the pooled expansion of combined power systems. The results of the study were released in October 1978
in a report entitled “An Evaluation of Strengthened Interprovincial Interconnections of Electric
Power Systems”.‘ In general, the report concluded that:
The putting in place of high-capacity electric transmission interconnections between the provinces
at the earliest feasible date would create the infrastructure required for electric energy to contribute
fully tothe objectives of a national energy strategy.
Andthat:
Within the scope of the examination made in this study, the prospective advantages of a strength-
ened interprovincial network merit further consideration.
Interconnections with Other Systems 65
And that:
Satisfactory arrangements for the establishment of a network of strengthened interprovincial in-
terconnections can be made provided that all parties have a desire to find a mutually satisfactory
solution and are willing to extend their cooperation tothat end.
The conceptual plan analysed by the IPACE study group calls for a total interprovincial transfer
capability of 13,650 MW by 2000. The existing capability is 1,330 MW and the additions under consid-
eration by the utilities would bring this to 6,130 MW by 2000. Thus the IPACE conceptual plan contem-
plates an additional capability of 7,520 MW. The corresponding data for interconnections with Ontario
are given in Table 5.3. The foregoing figures exclude the transmission capability dedicated to bringing
power from Churchill Falls in Labrador to load centres in Quebec and the transfer capability that can
be achieved by isolating certain generating stations (see Appendix C).
Table 5.3 Interconnections between Ontario and Its Neighbouring Provinces
Ontario-Quebec? Ontario-Manitoba
Existing (MW) 0 260
Under consideration by utilities (MW) 2,000 HVDC 260
IPACE plan (MW) 2,000 HVDC 2,000 HVDC
Total (MW) 4,000 3,520
Note a) Does not include the ability to isolate up to 1,300 MW of certain Quebec generation for the Ontario system.
Source: IPACE Study, vol. 1, pp. 13-14.
An economic analysis of the conceptual plan indicates a modest net economic advantage, particularly if
fuel prices continue to escalate relative to other costs. The economic benefit is realized by the reduction
in expenditures on generating capacity due to load diversity and the sharing of generation reserves,
and by fuel cost savings through co-ordinated operation resulting in better utilization of hydraulic and
nuclear resources. The report also noted: :
Other benefits provided by the plan but to which no dollars were attached include opportunities to
develop power projects on a regional basis, increased security of electricity supply, enhanced flexibil-
ity to adapt to changing fuel supplies and electricity demands, and flexibility of response to changing
energy policies.
The IPACE study proposes the formation of an interprovincial power co-ordinating council (IPCC) to
promote co-ordinated operation and planning of the provincial utility systems over strengthened
interconnections. IPCC would make the basic arrangements for financing, constructing, and operat-
ing the systems, for ratification by the provinces prior to implementation. To avoid all uncertainty as to
constitutional jurisdiction, IPCC should operate with authority delegated from the federal govern-
ment as well as from the provinces.
IPCC could operate the entire system as an interprovincial power pool in which the utilities would
participate in accordance with the provisions of an interprovincial power pool agreement. The inter-
provincial power pool would function in a manner similar to other power pools, in the United States and
overseas, and would offer flexibility of options tothe participants. Operation of the installed generating
capacity on the provincial utility systems would be co-ordinated so as to minimize the total cost of
electricity production consistent with safe, reliable operation and with environmental and other con-
straints. Mechanisms would be provided for co-ordinated planning of provincial generation pro-
grammes in a manner that would ensure that the basic responsibility for these development pro-
grammes would remain with the individual provinces. Each province would continue to plan and
execute its own generation development programmes. The IPCC would review these plans periodically
to identify opportunities for provinces to realize economies through the development of joint projects
and, with the concurrence of the provincial utilities, to make arrangements that define the extent and
duration of the participation by each party. The pool agreement would also specify the amounts of
reserve generating capacity to be carried by the provincial utilities and the ways in which it would be
shared in emergencies.
In order to establish major interprovincial interconnections, it will be necessary to resolve a number of
financial matters that are of critical importance to the success of the undertaking. In particular, it will
be necessary to determine an appropriate allocation of costs and benefits among the participating
governments and utilities, including the interrelated matters of the financing of the requisite capital
works and the pricing of interprovincial transfers of electricity. A number of considerations suggest
66 The Electric Power System
that it may be appropriate that governments assist in supplying the funds in excess of those required
for independent utility developments. The future stream of benefits over the life of the strengthened
interprovincial interconnections is uncertain, and some major benefits, such as improved reliability,
cannot be estimated in financial terms. There may be no direct relationship between the locations of the
electric power facilities and the regions where benefits are realized. A utility should not be expected to
provide funding except for use in its own province and to the extent that benefits are assured to it. It
appears appropriate that the balance of the funding should be provided by governments, for these
reasons and in order to expedite the installation of strengthened interconnections anticipating the
realization of broader provincial and national benefits.
The IPACKE study is already more than a year old. The data and load forecasts for the conceptual plan
were provided by the provincial utilities and were based on expectations in late 1977. The load forecasts
and planned expansion programmes have subsequently been reduced, at least in Ontario. This will
have some impact on the estimates of the conceptual plan. However, the purpose of the conceptual plan
was simply to raise possibilities and suggest preliminary conclusions. It has served that purpose well.
The implications for Ontario of the IPACE conceptual plan deserve further consideration.
Interconnections within Ontario
Ontario Hydro is the predominant producer of electric power in Ontario, but it is not the only one.
Statistics for 1976 indicate that 95 per cent of the generation is owned by Ontario Hydro or under its
direct operational control. Other utilities own 351M W or 1.6 per cent. Most of this is owned by the Great
Lakes Power Corporation (202 MW), the Canadian Niagara Power Company (95 MW), and the Gana-
noque Light and Water Company (11 MW). About a dozen municipal electricity utilities, mostly mem-
bers of the Ontario Hydro family, own 48 MW. Industries own 820 MW or 3.6 per cent of the provincial
total. This includes about 10 with 300 MW of hydraulic generation and about 20 with 520 MW of fossil-
fuelled generation, used mainly for the joint production of electricity and steam. A few of these indus-
tries operate small distributing utilities.
Most of the small Ontario producers augment their supplies with power purchased from Ontario Hydro
under the standard municipal electric or direct industrial tariffs. They may also purchase stand-by
service from Ontario Hydro to safeguard against the unavailability of their own generation. Indeed,
Ontario Hydro’s posted prices provide the criterion against which prospective producers determine
whether it is economical to install and operate their own production plants. This creates an institutional
bias that may discourage small producers, if the posted prices are lower than their marginal internal
costs.
Ontario Hydro purchases electricity from some small producers, such as the Mattawa Electric Light
and Power Company and the Great Lakes Power Corporation. In 1977, these purchases from Ontario
suppliers amounted to 464 GW-h, or 0.45 per cent, of Ontario Hydro’s total resources.®
On occasion, Ontario Hydro wheels power for the small producers, for example, from the Canadian
Niagara Power Company to the St. Lawrence Power Company, both of which are subsidiaries of the
Niagara Mohawk Power Corporation of New York State. In a recent move, Ontario Hydro has under-
taken to wheel power for Dow Chemical of Canada Limited from Sarnia to the international border, for
forwarding by the Michigan utilities to Dow Chemical in Michigan.
Inthe United States, the role of the small municipal electricity producers has been something of a cause
celébre. Large investor-owned utilities have been loath to interconnect and co-ordinate with small
public municipal utilities and rural co-operatives. The benefits often appear substantial to the smaller
party but trivial to the larger one. Furthermore, the larger tax-paying utility may be reluctant to share
its economies of scale with the smaller tax-free or even tax-supported utility. When such matters are
referred to it, the Federal Energy Regulatory Commission (FERC) has generally ruled that the small
utility is entitled to share the benefits of co-ordination, and the U.S. Supreme Court has upheld the
FERC on appeal.
Further examination of the role of the small power producer in Ontario would be desirable. An Ontario
power pool could be formed, to provide positive encouragement for small power producers to contribute
to the total provincial stock of generating capacity.
Interconnections with Other Systems 67
Conclusions
Ontario Hydro must continue to supply most of Ontario’s requirements for electricity from generating
stations situated within Ontario, and strategically located with respect to Ontario load centres. Techni-
cal, economic, social, and political factors all support Hydro’s role as Ontario’s chief producer of electric-
ity, mainly for use in Ontario.
Ontario Hydro has conducted a substantial electricity trade with neighbouring provinces and states. A
number of considerations suggest the desirability of increasing this trade, not only in absolute terms,
but also in terms of its share of the total provincial demand. Escalating fuel and capacity costs have
materially increased the value of intersystem transfers. Such transfers can promote the fuller utili-
zation of renewable hydroelectric resources and the conservation of dwindling fossil fuels. Strong
interconnections facilitate a flexible response to an uncertain energy future.
Both international electricity trade with the United States and interprovincial trade with other prov-
inces have been beneficial in the past and are expected to be beneficial in the future. The one should not
be pursued to the exclusion of the other, nor should the one be delayed on account of the other. Reason-
able goals for implementation in the 1990s might be the provision and utilization of transfer capabili-
ties of some 2,000 to 4,000 MW between Ontario Hydro’s East System and each of the following:
Quebec; the West System and Manitoba; and New York, Michigan, and Ohio.
Power transfer capability with the United States is projected to decline sharply as internal transmis-
sion in southern Ontario becomes fully dedicated to provincial loads. To maintain and increase interna-
tional electricity trade will require the construction of major new transmission facilities. The planning
of these facilities will be a complex task — one that must take cognizance of existing and projected
power system configurations in Ontario, the border states, and beyond, as well as of alternative land
uses and socio-environmental impacts.
While existing interconnection agreements with U.S. utilities apparently serve Ontario’s interests
well, it will be necessary to review them critically during the planning of this new transmission.
In many respects the strengthening of interprovincial interconnections is more challenging than the
strengthening of international interconnections. Strong interconnections with Quebec and Manitoba
and their possible extension farther east and west would probably require comprehensive agreements,
between the provincial and federal governments as well as between the electricity utilities. The era of
firm interprovincial power transfers has ended and must be supplanted by new understandings, possi-
bly incorporating pooling or co-ordination concepts. Reasonable compromises must be sought to accom-
modate both north-south and east-west transactions.
Ontario Hydro has negotiated mutually beneficial interconnection agreements with utilities in neigh-
bouring states and provinces in the past and is expected to continue to do so in the future, within the
framework of jurisdictional constraints. It is desirable to study the implications of Hydro’s applying
similar methods in its dealings over interconnections with small utilities and industrial producers of
electricity within Ontario.
68 The Electric Power System
Fig. 6.1: p. 76
CHAPTER SIX
Operation and Control of the System
The operation of a large power system on a minute-by-minute, day-by-day basis is predicated on
matching the generation to the load by making the best use of the available power resources (including
assistance over the interconnections), while ensuring an adequate level of reliability and security of the
system. The amount of generation that is fed into the system at any instant is exactly equal to the load
that is being supplied. This load includes the transmission losses that are associated with the use of the
electrical equipment on the system at the time. Generating capacity is added to or taken off the line
according to the hourly customer load and scheduled outages or deratings of generating units, Allow-
ance is also made for unscheduled shut-downs of generating units through the so-called “spinning
reserve” (generating capacity that is operating, but not loaded to its maximum output).
In matching the generation to the load, a number of alternatives are usually available, in the numbers
and types of generating units that can be operated at any given time. The choice is governed by the
criteria of maintaining a specified level of system reliability and security (that is, minimizing both the
likelihood of a breakdown of supply and the severity of any interruption that does occur) and minimiz-
ing the cost of generation. While the possibility of an interruption of supply can never be entirely
eliminated, proper system operation procedures can reduce the chance toa minimum.
The Operating Control of a Power System
The operation of a large power system such as Ontario Hydro’s is performed by a three-tiered structure
of operators (Figure 6. 1).
At the lowest level are the station operators, their assistants, and their agents, who control the distri-
bution system feeding power from the subtransmission system tothe large direct customers, and to the
retail distribution systems; and operate, under direction, generating stations and portions of the
subtransmission system.
The middle tier of operators are the regional operators and their assistants, who control the subtrans-
mission system within a region of the system; and operate, under direction, portions of the main
transmission system.
The highest of the three tiers of operators are the operators and schedulers who man the system’s main
control centre, for example Ontario Hydro’s Richview Control Centre, where they direct the operation
of the generating stations, the interconnections with other systems, and the main transmission sys-
tem, which are collectively called the “bulk power system”’.
The limits of responsibility for each tier of operators are determined mainly by the operational signifi-
eance of the various elements of the system rather than by generator capacities, transmission line
voltages, or power flows. All generating and transmission equipment that can have a significant effect
on the security or economy of the system as a whole is considered to be within the bulk power system’s
boundaries.
Control of the operation of the bulk power system involves the authorization of the switching of trans-
mission lines and other high-voltage apparatus within the system, the control of voltage levels, the
monitoring of power flows and equipment status, the avoidance of security limits, the authorization of
any work on or adjacent to the system’s high-voltage apparatus and its associated protection and
control apparatus, the authorization of all generation and transmission outages, and the scheduling
(and subsequent loading) of all generation on the system and of interconnections with neighbouring
systems.
The system control centre’s operators and schedulers make their normal day-by-day and minute-by-
minute decisions according to policies, plans, and directives developed by groups in the power system
organization that plan the operation of the system. Policies, plans, and directives governing the two
lower tiers of operations are developed by the head office, and by regional and local operations groups. It
is the quality of these policies, plans, and directives (and their faithful execution), together with the
effectiveness of the responses by all levels of operators to unplanned events, that largely determine the
economy, security, and reliability with which the system operates.
The foregoing assumes that adequate protection, indication, control, and communication devices are
Operation and Control of the System 69
provided at all levels of the system. The correct functioning of the automatic protection and control
equipment largely determines the system’s appropriate response to routine changes, unexpected mal-
function and failure of equipment, and actions of the environment.
Protection, Indication, Control, and Communications
Protection
Each item of high-voltage equipment is subject to failure. A failure may be the result of a “fault”, that
is, an uncontrolled discharge of power between high-voltage conductors and/or between high-voltage
conductors and ground. The magnitude of the electric current that is involved in a system fault is such
that the system’s equipment and public property may be severely damaged and human lives endan-
gered if the fault is prolonged. In addition, system faults can cause instability in the bulk power
system. Some examples of failures involving system faults are:
e treelimbs blown against transmission-line conductors
e insulator ‘“flashover” resulting from alightning stroke
e failureoftheinsulation ina generator cable, transformer, or circuit-breaker
High-voltage equipment can also suffer failures that do not (at least initially) result in system faults.
These failures, however, can result in severe damage to equipment and danger to employees and the
public, and can ultimately result in system faults. Examples of such failures are:
e overheating of transformer windings
e excessive vibration in turbine-generator sets
e failureof boiler orturbine auxiliary equipment
In order to prevent damage to equipment and danger to the employees and public, each piece of high-
voltage equipment in the system is provided with low-voltage apparatus to detect failures and faults
and initiate the disconnecting of the equipment from the system through the opening of circuit-
breakers. The functioning of this automatic protection apparatus is designed to remove the faulty
equipment as quickly as possible, while avoiding unnecessary removal of the system’s other high-
voltage equipment.
Because of the complex nature of the system’s power flows, of the protection equipment, and of the
switchgear, it is not possible to guarantee that faults and failures will be correctly detected and acted
upon. Also, there is always the possibility that protection equipment and switchgear will malfunction,
resulting either in a failure to disconnect the faulted equipment from the system or in the removal of
too much equipment. To guard against the the failure to remove a fault, protection schemes are dupli-
cated and back-up schemes are provided. From a system point of view, the most severe fault conditions
occur at higher system voltages and close to sources of generation. Therefore, it is usual to find the more
elaborate and speedier protection schemes in such locations.
The portions of the system that are the easiest to protect are those whose function is simply to feed
power to load areas, and where, consequently, it is relatively easy to discriminate between load currents
and fault currents. However, protection for these portions of the system can be greatly complicated if a
number of small generating units are added near the load centres. These units, when mingled with the
load, can reverse or alter the direction and magnitude of power flows, which makes it difficult to
distinguish between load currents and fault currents.
Indication
At all levels, the system’s operators are provided with a variety of indications of the condition of the
portion of the system for which they are responsible. Indications include:
e circuit-breaker and switch positions (open or closed)
e voltage measurements
e realandreactive power flows
e generating unitoutputs
e transformertapchanger positions
In addition to this type of continuous indication, alarms are provided to alert the operators to such
events as:
e theopening of acircuit-breaker by automatic protection apparatus
e ahigh-orlow-voltage condition
70 The Electric Power System
e the failure of auxiliary apparatus
e abnormal conditions within high-voltage apparatus (e.g., high temperatures in transformer
winding)
© theexistence of an undefined abnormal condition at a minor location
All of these indications are necessary, so that the operators are continuously aware of the state of the
system and particularly of the portion over which they are exercising control. The operators require
this knowledge of the state of the system, so that they can:
e make the alterations that are necessary from time to time to maintain the system’s voltage and
power-flow conditions within acceptable security limits
e arrange the pattern of generation and interchange in the most economic fashion
° ae appropriate action to safeguard the system, and prevent damage to equipment or danger to
people
e restore the system to a normal, secure operating state following a disturbance of the sort that
could be producedbyafault —
At some locations in the system, where there is too much information for a single human mind to
assimilate, computers are used to process the information and give the operators a concise picture of the
state of the system.
Control
The power system and its operators are provided with a considerable amount of control equipment for
use in operating the system. Controls may be automatic or manual, or both. However, all automatic
controls are capable of being overridden at the discretion of the operators.
Suitable controls are provided for all items of equipment that are important to the functioning of the
system. The controls are too numerous tolist in full, but some examples are given below.
e Generators are provided with automatic excitation control so that a desired level of output volt-
age can be maintained under varying output conditions.
e Many transformers are provided with automatically controlled tap changers so that a constant
secondary voltage can be maintained under the changing levels of load being supplied through the
transformers.
e Many transmission line circuit-breakers have “auto-reclosing” controls that permit the circuit-
breakers to be automatically reclosed after a protection operation has been carried out to clear a
fault.
e Many circuit-breakers have remote controls that allow them to be operated from central control
locations. Most important among such circuit-breakers are those that are used in conjunction with
transmission lines, transformers, customer feeders, and static capacitor banks. The circuit-
breaker is the most-used type of control in the system.
e Load, frequency, and tie-line controls are provided at the overall system level to automatically
and continuously adjust the output of selected generation units or stations in the most economic
fashion to maintain pre-set system export or import, while at the same time responding to fre-
quency variations in the system.
e Generating unit remote controls are used for many remote hydraulic generating stations, to
permit operators at central locations to start up, shut down, and control the output of the
generators.
e Special system controls are sometimes provided to take care of difficult system situations, in
which security limits are being violated and unusual corrective action must be taken so rapidly that
there is not time for human response. Ontario’s power system has situations such as this in the
transport of power out of the Bruce nuclear complex and in the maintaining of adequate voltage in
the Ottawa area.
e Loadmanagementcontrolsare installed in many systems to permit operators at central locations
to switch various end-use devices such as electric hot-water heaters on and off in consumers’
premises.
All system elements capable of being operated are, of course, provided with local controls to permit
operation.
Operation and Control of the System 71
Communications
The operation and control of a large power system relies heavily on communications. Verbal informa-
tion and instruction has to be passed between operators at various locations in the system. Information
to provide system indications is passed from source to destination over great distances. Control signals
are also required, and protection signals and information must often be passed from one end of a
transmission line tothe other.
All of this communication activity requires a great deal of communication apparatus. Power systems
use a variety of media including:
e telephone lines (dedicated or shared)
e privatecommunication cables
e power-line carrier equipment (which uses the transmission lines to carry communication
signals)
e point-to-point microwave radio
e mobile microwave radio
In recent years, communications have become so important to the operation of a large power system
that it is necessary to provide a system of alarms and indications to exhibit the state of the communica-
tions network. Also, for most functions, to safeguard their security it is necessary to provide more than
one communications path. The communications network for a large power system can become a system
in itself, requiring monitoring and control. Ontario Hydro hasa central microwave control room where
shift personnel carry on a full-time control function.
The Management of a Bulk Power System
The management of a bulk power system, and its operation, is performed according to criteria that are
mainly self-imposed, but derived from industry standards that have developed along with the industry
itself. Perhaps the one major exception to this is the area of environmental protection, which is subject
to government regulation. Overall, the criteria relate to:
e theneedtosupply alloftheload
e economy
security
reliability
the safety of the public, of utility personnel, and of utility equipment
the environment
Usually, the power system is managed and operated in such a way as to supply the load with the
maximum economy possible without violating standards relating to security, reliability, safety, and
the protection of the environment.
The management of a bulk power system is carried out by head office groups that develop policies, plans,
and procedures, and analyse all available data to ensure that the system adheres to established design
criteria at minimum cost and is able to respond to emergencies, including those in which events cause
the design criteria to be exceeded. Specifically, the management of a bulk power system involves:
e developing plans for the utilization of all generating resources and interconnections with other
utilities
e developing operating standards and training programmes
e analysing the behaviour of the system on the basis of actual performance, and through transient-
analysis and load-flow programmes defining operating limits under various modes of operation
e making recommendations to other work groups concerning the operating acceptability of major
alterations or additions tothe system
e establishing and monitoring standards for protection, indication, metering, control, and com-
munication apparatus
e analysing the performance of protection equipment and taking the action necessary to correct
sub-standard performance
e collecting and analysing relevant statistical data on loads, resources, fuels, stream flows, water
levels, and storages
Generally, the management and operation of a bulk power system are divided between electrical
operations aspects and economic operations aspects, together with provision for co-ordination at all
management and operational levels.
72 The Electric Power System
Electrical Operation of the Power System
In simple terms, the management and operation of the electrical aspects of the system are aimed at
ensuring as far as possible that the system’s loads remain connected and supplied, that the transport
capability does not hinder the economic operation of the system, that customer voltages are within
correct limits, and that the system does not present a hazard to the public or to the utility’s employees. It
is also required that equipment be maintained and augmented.
To achieve these aims, the activities of the head office management groups involve:
e developing policies and general instructions for the operators
e establishing limits for the maximum power flows at various points in the system under normal
and abnormal conditions in order to prevent equipment overloading and system instability prior
to, during, or after faulting or other loss of equipment
e providing correct settings for voltage-regulating equipment
e planning major maintenanee and construction outages of transmission equipment
e arranging for special system connections or special operating procedures to cover particular
abnormal situations
e analysing theelectrical performance of thesystem
The main computational aids for this work by head office groups are load-flow and transient-stability
computer programmes.
The activities of control centre schedulers and other members of the electrical operations staff involve:
e scheduling and co-ordinating the outages of high-voltage equipment and the protection, commu-
nication, and control apparatus
° processing information and instructions from the head office groups into a form suitable for the
operators
e providing locally developed instructions tothe operators
e inputting information and limits intocontrol-centre computer systems
e analysing the results of system operation and passing information tothe head office groups
The operators at the various levels of operational control are the people with actual physical control of
the system. While the activities of these operators include the continuous carrying out of the instruc-
tions and plans provided to them, they also include the continuous monitoring and correction of volt-
ages and power flows. Also, the operators perform the prearranged switching and isolation that is
required to render equipment safe to work on. One of the most critical activities of the operators is to
respond to sudden abnormal conditions brought about by faults and other malfunctions. The response
may require the simple reclosing of circuit-breakers after a temporary fault, the isolation of certain
items of equipment and the arranging of their emergency repairs, or the re-connecting of a major
portion of the system after a serious disturbance. At high levels of operational control, the operators
may be assisted by computer systems that digest system-state information, monitor limits, analyse
possible contingencies, and perhaps even suggest corrective strategies.
The greater the number of restrictive system limits (usually brought about by outages or inadequate
transmission equipment), the more difficult it is for operators to maintain the system in a secure state
with sufficient scope for economic operation, and the more frequent are the occurrences of operator
errors.
The Economic Operation of the System
The economic management and operation of the system does not involve as many people as does its
electrical operation. Those involved are, primarily, the head office groups that plan system production,
the production schedulers and shift operators at the main control centre, and the station operators who
control the generating units.
In discussing the function of the head office groups, it is useful to consider the variables that can affect
the economic and reliable operation of a generating system.
There are a number of independentor uncontrolled variables:
e system demands
e forced generator outages
e hydraulicinflows
e environmental conditions
Operation and Control of the System 738
e conditionson neighbouring systems
e interruptionsto fuel supplies
In coping with the uncertainties that are implied by these uncontrolled variables, the production
planners can control a number of other variables:
e generator maintenance schedules
e hydraulicstorage discharge strategies
e interchangeagreementsand contracts with neighbouring systems
fuel-supply contracts and fuel stockpiles
It is the function of the production planners to produce a co-ordinated set of maintenance schedules,
hydraulic resource release plans, interchange contracts, and fuel-supply and fuel-use strategies that
can be used to minimize overall production costs while respecting planning criteria for generation and
fuel-supply reliability as well as environmental and system security constraints. To produce such a set
of plans, the planners make use of sophisticated, computerized mathematical models, as well as of their
own judgement based on experience.
The nature of the mathematical techniques that can be applied to these problems is such that the
difficulty of arriving at the most economical set of plans increases with any increase in:
the number of generating units
the number of types of fuel used in thesystem
fuel constraints
dual-purpose applications (for example, steam and electricity)
environmental constraints
transmission limitations
The mathematical models must take adequate account of the physical characteristics of the generating
units (the time taken to start them up and shut them down, their ability to change output levels, the
change in production cost or water use associated with a change in output level, etc.).
All production plans do of course, exhibit certain predictable characteristics, for example:
e theincrease of planned maintenance during low-load months of the year
e the higher expected use of thermal generating units with lower incremental production costs,
compared with units with higher incremental costs
e theuseof hydraulic generation with good storage to displace the highest-cost thermal generation
The production plans must also take account of the availability of interruptible and managed loads and
attempt to optimize their use. Production plans are subject to continual change as new information
concerning the uncontrolled variables becomes available.
The production schedulers at the main control centre receive these plans and use them to provide
instructions to the operators for the hour-by-hour operation of the generating system. The main in-
structions to the operators are contained in daily and weekly production schedules. The weekly sched-
ule is revised as often as necessary for the next seven-day period. The production schedule is a set of
hourly output levels for each generating unit or station on the system, together with hourly values of
power transfer with neighbouring systems at each interchange point. The production schedule is based
on up-to-date forecasts of hourly loads, generator availability, environmental conditions, and hydrau-
lic levels and inflows, and it attempts to optimize the use of all available resources including pumped
storage, managed loads, interruptible loads, and interchange possibilities, while respecting equip-
ment, fuel, and hydraulic system security and environmental constraints.
Generally, the production schedule is prepared with the aid of a sophisticated computer programme
that contains a very detailed mathematical model of the generating resources of the system and takes
account of a number of requirements, such as the amount and distribution of operating reserve (the
spare generating capacity that can be loaded quickly in emergencies). Again, the creation of the
production schedule is complicated by any increase in the number of generators and constraints.
There is a trend towards the use of computer models to provide inputs to the production scheduling
process — such as load forecasts and hydraulic inflow forecasts. There is also a trend towards installa-
tion of production scheduling programmes on control-centre computer systems, so that updated sched-
ules can be produced from the latest information gathered by the control computers, at any time the
system operators desire.
74 The Electric Power System
The control-centre operators who are responsible for hour-by-hour system production use the produc-
tion schedule to arrange the loading of generating units and to arrange economic interchanges of
power and energy with neighbouring systems. These operators also place suitable generating units, as
defined by the schedule, on automatic load control to respond to minute-to-minute and second-to-second
variations in system load. They are also responsible for making emergency changes in unit loadings
and interchange flows, to compensate for any loss of system generation equipment or transmission
capability on their own or on the neighbouring systems.
The overall results of the system’s economic operation exhibit certain predictable characteristics, such
as:
e the “stacking”, or “merit order loading”, of thermal generation, so that the units with the lowest
incremental cost are run for the longest periods
e the “shaving” of peak loads by using hydraulic and thermal generating units that have limited
energy capability
e generation by pumped storage plants during peak-load hours, and the pumping of water into
storage during low-load hours
e thereserve capacity maintained through the units that have rapid response characteristics
e the reduction overnight to minimum load of large thermal units that take a long time to shut
down and start up
The actual production on the system is analysed by the schedulers and production planners whose job it
is to identify correctable problems, that have prevented the system from achieving its economic
optimum.
Operation and Control of the System 75
si0jesado uoijeys — waysAs uoIyNqisjsiq
Sioyeiado jeuol8ai — waySAs UOIssIWSUeLGNS
sia|npayas pue suojeiado aquad joxju0d waysés — wajsAs samod ying
See
SJaWO}sn}
VW Ww
iam
Saul UOISSIWSUEL} Jamod yIng
ddJ0Y “20sN0S
SwWaSAS Jay}
Wa}SAS JaMOg & 40 {O4JUO) jeUOI}eJedQ 40 Sjana] [°g “INBIy
76 The Electric Power System
CHAPTER SEVEN
The Planning of the Electric Power System in Ontario
The purpose of an electric power system is to supply a geographical region with the electric power and
energy needed for its homes, industries, farms, and institutions — reliably and at “reasonable cost”.
The planner of such a system has the task of defining as explicitly as possible the optimal strategy for its
development, so that the costs and the benefits of a service of the desired quality are in balance. The
questions the planner must address, and find answers for in the context of the new generation, trans-
mission, and distribution facilities, are: What type? How much? When? and Where?
The Changing Planning Environment
A basic input to the process of planning is the demand for electric power and energy. From 1935 until
recently, the demand for electric energy in Ontario grew at a rate of about 7 per cent per year. Both the
growth in demand and the retiring of old facilities due to ageing or obsolescence require that new
facilities be added to the system. Planning in today’s climate for the addition of new facilities is a
challenging task because of the long lead time associated with major facilities and the uncertainty
concerning the growth in demand over the lead-time period.
The total lead time for a major generating facility is between 10 and 15 years (see Chapter 3), and fora
major transmission line it is about eight years. Before the OPEC oil embargo in 1973, the growth in the
electricity utility industry in North America was quite regular, and the forecasts over a 10-year period
were reasonably accurate. In Ontario Hydro’s case, the forecasting errors within a 10-year period were
generally less than 10 per cent.!
In the days of steady growth in demand and short lead times, the task of system planning was compar-
atively simple. The inflation rate was low, fuel supply and sound financing were regarded as assured,
and cost estimates were assumed by planners to be reliable. The environmental impact of electric power
production and supply had not yet become a social issue, and thus the acquisition of land for generating
stations and transmission rights of way was not a major problem. The load forecaster gave his projec-
tions to the system planner, whose time was spent mainly on immediate capacity additions rather than
on system plans for 15 or 20 years ahead.
However, this situation changed dramatically in the 1970s. The forecast made in 1969 for peak demand
in Ontario in 1978 was about 30 per cent higher than the demand turned out to be. Ontario Hydro’s
forecast of the average annual growth rate for demand in the 1980s has dropped steadily from about
seven per cent in 1975 to 4.7 per cent in 1979. Some experts believe that even this forecast is too high
and predict a 3 per cent annual rate of growth.
Other factors that have altered many of the traditional concepts of system planning are: the impact of
alternative technologies, uncertainty concerning fuel and capital availability, and unforeseen plant
shut-downs. Alternative technologies — co-generation, heat pumps, solar heating, biomass — will have
their influence, but their precise eventual role is difficult to foresee. Most large-scale conventional
generating plants cost hundreds of millions of dollars (the estimated cost of Pickering B, coming into
service in 1981, is about $2 billion. The size of the investment, combined with the long lead times, makes
the financing of such plants a major undertaking. Before a plant is built, the supply of fuel over the
plant’s service life (normally 30 years) must be reasonably assured. Plants may be shut down for
extended periods by regulatory agencies of the government or by the utility itself, for safety or en-
vironmental reasons. For example, Ontario Hydro recently announced that it plans to take the four
Pickering A units and the first three Bruce A units out of service for one year each starting in 1985, to
replace pressure tubes that have stretched more than expected. In March 1979, the U.S. Nuclear Reg-
ulatory Commission (NRC) ordered the shutting down of five nuclear power plants in the northeastern
U.S. because of doubts about their ability to withstand earthquakes. The NRC ordered the shut-downs
after the firm that designed all five plants discovered a mathematical error in the computer pro-
grammes that were used to design some of the plants’ cooling systems.
These uncertainties may culminate in either an excess or a shortfall in the capacity of the electric power
system. In either case, the result will be to increase the cost of power to the customer. The largest single
component of the cost of excess capacity is the interest on the capital borrowed to finance the construc-
tion of the capacity. Other charges, such as depreciation and operations and maintenance costs, depend
The Planning of the Electric Power System in Ontario 77
on whether the capacity is in service or “‘mothballed”. It is possible to offset the cost of excess capacity by
finding export markets. At present, Ontario Hydro has an excess capacity of about 3,400 MW (that is,
3,400 MW in excess of the 25 per cent generating reserve margin required over the annual firm peak
load). On the basis of the capital cost of units 5 to 8 of the Nanticoke Generating Station, the cost of this
excess capacity is approximately $100 million a year. Hydro is pursuing the exporting of firm power to
U.S. utilities in order to reduce this surplus and has had some success (see the section on the outlook for
electricity trade in Chapter 5).
The cost of a shortfall in capacity is the loss sustained by customers due to interruptions of supply. The
cost of a shortfall depends on many factors; important among them are the type of customer (indus-
trial, commercial, or residential), the size, frequency, and duration of the interruption, and whether the
customer receives sufficient advance notification. Ontario Hydro’s estimates of the cost of interruption,
based on a survey of large industrial users, range from $0 to $91/kW for a one-hour interruption.” A
shortfall in system capacity does not always result in an interruption of service. Hydro and many North
American utilities are interconnected with their neighbours, and in an emergency a utility may be able
to purchase power to minimize interruptions. The price of the purchased power may be quite high,
however.
Having discussed the factors crucial to planning in today’s environment, we will outline the basic
methodology used by the electricity utilities for system planning and then look into Ontario Hydro’s
practice and policies with respect to planning its electric power system. Since the responsibility for the
distribution of electric power in Ontario lies toa large extent with municipal electricity utilities, we will
focus our attention on planning for generation and bulk power transmission.
The General Approach to Planning Generation
The first element in the methodology for planning the addition of new generating facilities is a load
forecast for 15 to 20 years ahead. This forecast includes not only the peak demand but also the amount
of energy that will have to be supplied from year to year and the possible changes in future load
characteristics.
In determining the amount of new generating capacity that will be needed, two more factors are
significant. In order to meet the peak-load conditions in some future year at a specified level of reliabil-
ity, a certain amount of reserve capacity is required for protection against forced and scheduled plant
shut-downs and deratings (see Chapter 4). While the actual amount of reserve capacity that will be
required depends on the configuration of the system, an initial estimate may be made in the form of a
percentage based on experience and judgement, e.g., 25 per cent for the Ontario Hydro system. In
addition, it will be necessary to replace existing capacity after it has completed its useful life. Thus, the
expected growth in load, the associated reserve margin, and an allowance for plant retirement provide
the system planner with an estimate of the total amount of new generating capacity that will be
required over the planning period.
The next step is to determine which of the alternative forms of generation should be used to meet the
new capacity requirement. This involves an economic study of these alternatives to determine their
suitability for supplying loads of various durations, taking into account the operating characteristics
of the alternatives. With this information, along with the data on generating capacity, existing and
under construction, the system planner is able to formulate some broad guidelines for the expansion of
the generating system. Starting with these guidelines, the planner develops a number of feasible
alternative plans with a view to discovering the “optimum” plan. Considerations that may restrict the
number of feasible alternatives are capital, fuel, site and manpower availability, and environmental
impact. The process of determining the “optimum” plan is clearly iterative. Each alternative plan is
evaluated in conjunction with the load forecast, to obtain estimates of capital requirements, fuel re-
quirements, operating costs, reliability, manpower requirements, etc. The plan with the minimum cost
that also satisfies the reliability and other constraints is selected as the basis for system expansion.
78 The Electric Power System
Ontario Hydro’s Long-Range Generation Forecasts
Ontario Hydro, like many other utilities, plans its future development of generation resources to meet
the forecast growth in load at a given level of reliability. These plans are subject to the constraints of
capital and fuel availability, environmental impact, and safety. Between 1974 and 1977, Hydro pro-
posed three basic long-range forecasts (LRFs) in the face of changing load forecasts and the con-
straints on borrowing imposed by the government. These are identified as LRF 41, LRF 48, and LRF
48. In January 1975, Hydro’s Board of Directors selected LRF 414A, a variation of plan 41, for use in
hearings before the Ontario Energy Board. Soon after that, LRF 43 was developed to conform to the
1975 load forecast.
In July 1975, the Treasurer of Ontario asked Ontario Hydro to reduce its capital expenditures to 1985
by $1 billion. In response to this request, Hydro modified its plans and proposed LRF 48P (Table 7.1).
The 1976 load forecast indicated a slight drop in the expected growth in load. Hydro was therefore
asked by the Treasurer to limit its eapital borrowings in 1976, 1977, and 1978 to $1.5 billion each year.
Later in 1976, Hydro modified its 1976 forecast to allow for the likely impact of load-management and
conservation programmes. To incorporate these changes, Hydro produced LRF 48 (Table 7.1). How-
ever, the load forecasts continued their downward trend, and, consequently, LRF 48 was modified
slightly (LRF 48A in Table 7.1) to conform tothe 1977 load forecast. The first major departure from the
historical exponential load growth at 6.5 to 7 per cent annually was evident in the 1978 load forecast,
which projected an annual growth rate of 5.4 per cent over the next two decades. A considerably
smaller, yet still large, generation plan called “Program Z” was proposed in response to the 1978 load
forecast (Table 7.1). Hydro’s 1979 forecast indicated yet another significant drop in load growth (4.5
per cent per annum to 2000). We will discuss Hydro’s current generation programme in a later section.
Table 7.1 Ontario Hydro East System — Load and Generation Forecasts (1980-95)
Load forecast 1975 1976 1977 1978
Average annual growth rate (%) 7.05 6.85 6.4 5.4
LRF 43P 48 48A ay i
1995 Primary peak (MW) 57,203 52,020 48,492 38,182
1995 Generating mix (MW)
oil and gas 5,425 5,425 5,425 4,325
(%) 7.6 8.8 9 9.1
coal 22,562 19,562 19,562 14,318
(%) 31.6 31.7 32.3 30.0
nuclear 37,714 30.984 29,856 23,348
(%) 52.8 50.2 49.3 48.9
hydraulic 5,710 5,710 5,710 5,710
(%) 8.0 9.3 9.4 12.0
1995 Total capacity (MW) UALS 61671 60,553 47,701
Reserve on primary peak (%) 25 19 25 25
Source: RCEPP.
Most of Ontario Hydro’s LRFs cover a period of 20 years. Hydro has stated that, because of the uncer-
tainties about future load and generation requirements, it is not reasonable to expect to be able to
devise a single, specific, fixed, year-by-year programme of new facilities for the next 20 years. Each
new project is authorized for design and construction only when that is essential. However, the LRF's
are necessary to set guidelines for the authorization of new projects and to ensure that each project,
once built, will be useful throughout its life. These guidelines relate to the projected nature, timing, and
amount of new generation capacity.
As with most utilities, Ontario Hydro’s primary basis for the selection of a long-range plan is the
minimization of economic costs. However, the nature of these costs has changed. Until recently, they
were the long-run costs to Hydro, but now there is more emphasis on the short-run cost of power to the
customer, reflecting the effects of raising funds for capital construction. As mentioned earlier,
between 1975 and 1977, the generation programmes were also constrained by the provincial borrow-
ing limits.
Although Ontario Hydro planners use sophisticated mathematical models to evaluate reliability and
economic costs of generation alternatives, they also use their own judgement in weighing qualitative
The Planning of the Electric Power System in Ontario 79
factors that are difficult to incorporate into mathematical models. These factors include the socio-
environmental constraints and the uncertainty associated with the dominant planning variables —
load forecasts, capital availability, lead times, fuel supplies, and the effect of load management and
conservation. The nature of Hydro’s long-range generation programmes has been described as
follows:°
The generation programs that you see are a best-guess forecast by Ontario Hydro or the planners at
any given instant as to what we think the most likely generation program will be. Itis primarily used
as an internal forecasting tool. We are not saying on a given set of rules, such as civil engineering
economics, that it is an optimum generation program, really, it is somewhat an optimum generation
program which includes all constraints that we can both quantify and study and the ones that we
cannot quantify.
Ontario Hydro, in 1976, presented to the Commission its basis for selecting the generation expansion
programme LRF 48.4 Below isa summary of the factors Hydro mentioned. There is no evidence to date
to indicate any significant departure from them.
e CANDU nuclear units should be used, as much as possible, for future base-load requirements.
e Future fossil-steam units should be based on coal, and major commitments to oil- or gas-fired
units should be avoided. These units, along with further hydraulic and energy-storage schemes,
should be used for reserve, peaking, and intermediate-load applications. Coal-fired generation
should also be used to supply the part of base load that is not supplied by nuclear or base-load
hydraulic units. .
e Increased reliance should be put on western Canadian coal, assuming that the cost is reasonable.
e Newthermal stations should be large and centrally located near large bodies of water. However,
smaller multi-purpose stations, which may become economic, may be located inland.
e Newelectricity generation technologies — solar, wind, geothermal, fusion, etc. — are not expected
to have a major impact until 2000.
e Development of the hydraulic potential of the rivers flowing into James Bay and Hudson Bay is
likely to be affected by economic, social, environmental, and political considerations.
e Purchases from neighbouring utilities should be undertaken when economic.
These broad planning concepts are appropriate as they apply to the reliability and operational aspects
of system design. However, there are also many economic, environmental, social, and political issues
related to these concepts, and they are discussed in other volumes of this Report. As indicated in
Chapter 3, the principles of flexibility of the planning process and resilient system design must be
associated with these broad concepts.
Anunderstanding of Ontario Hydro’s planning objectives with respect to generating mix and reliabil-
ity may be obtained by comparing Hydro’s long-range generation forecasts. Table 7.1 presents the key
elements of four long-range forecasts: LRF 43P, LRF 48, LRF 48A, and “Program Z”. (Hydro’s 1979
generation programme is discussed in a later section.) Between 1975 and 1978, the forecast average
annual growth rate for primary peak over the 1980-95 period declined from 7.05 per cent to 5.4 per
cent, resulting in a reduction of about 19,000 MW in the predicted primary peak for 1995, from 57,203
MW to 38,182 MW.
Table 7.1 shows that while the total planned capacity drops with the demand, the generating capacity
mix does not change by any significant amount. The total hydraulic capacity remains constant at the
current level, and, as the forecast load drops, its share inthe 1995 system increases from 8 per cent to 12
per cent. This increase in the hydraulic share is offset by a corresponding reduction in the share of
nuclear capacity. The shares of coal- and oil- and gas-fired generating capacity do not change greatly.
It is reasonable to conclude that, under various LRFs, the planned capacity mix for 1995 is approx-
imately 50 per cent nuclear, 30 per cent coal, 10 per cent hydroelectric, and 10 per cent oil and gas. As
will be seen later, this approximate capacity mix also applies to the 1979 generation forecast. It is
interesting to note that the 1977 forecast of the primary peak in 1995 is about 3,500 MW lower than the
1976 forecast, but the corresponding capacity reduction is only about 1,000 MW (Table 7.1). The main
purpose of this appears to be to restore the reserve margin (which was compromised because of capital
constraints) to the “pre-constraint” level.
Figures 7.1 and 7.2 illustrate how measures of generating mix and reliability vary between 1980 and
1995 under the four LRFs. Figure 7.1 shows the planned share of nuclear capacity as a percentage of
firm peak demand. A major impact of capital constraints on the generation programme is apparent by
comparing LRF 43P and LRF 48. The share of nuclear in LRF 48 is consistently 5 to8 per cent less than
re ee ee
80 The Electric Power System
Fig. 7.1: p
Fig. 7.2: p. 93
in LRF 48P beyond the mid 1980s. This caused a corresponding reduction in the reserve margin over
the same period (Figure 7.2).
The severity of the capital constraints imposed on LRF 43P and LRF 48 was reduced considerably
when LRF 48A was presented, with a much lower 1977 load forecast. The planned share of nuclear
capacity and the reserve margin both rose materially (Figures 7.1 and 7.2). In terms of their size,
generation programmes LRF 48 and LRF 48A were not very different (Table 7.1); LRF 48A has 1,000
MW less nuclear capacity than LRF 48. A key factor in the development of LRF 48A was the problem of
scheduling generation at sites owned by Ontario H ydro.
With the substantially reduced 1978 load forecast and corresponding ‘‘Program Z”, the long-term
effect of capital constraints disappeared. The nuclear capacity’s share of the firm peak load in 1995 is
about 65 per cent (Figure 7.1), and this, combined with about 2,800 MW of existing base-load hydroelec-
tric capacity, is close to Ontario Hydro’s base-load requirements. The planned reserve levels under
“Program Z” are also considerably higher (Figure 7 .2). The cause of excessive reserve margins
throughout the 1980s is that the committed nuclear programme was not deferred and the only reduc-
tion made to the fossil programme over this period was the cancellation of two of the four projected 547
MW oil-fired units at the Wesleyville Generating Station. The reasons given by Hydro for not defer-
ring the committed programme were the cost of electricity to the customers, employment, and the
likely impact on the provincial economy of aslow-down.
On the basis of the analysis presented in this section, the following observations may be made:
e Because of their short-to-intermediate-term nature, the capital constraints did not greatly affect
Ontario Hydro’s long-term generating mix.
e Achangein load forecast did not affect Hydro’slong-term generating mix.
e There was a significant impact on the planned system reserve margins as a result of capital
constraints.
e The generation programmes did not necessarily change as the load forecasts changed, but the
projected reliability standards changed.
e Ontario Hydro’s planned long-term mix of generating capacity in this century is 50 per cent
nuclear, 30 per cent coal, 10 per cent hydroelectric, and 10 per cent oil and gas. (It should be noted
that the share of existing and committed hydroelectric and oil- and gas-fired capacity may rise as
the load forecast drops. This may cause the share of coal-fired capacity to decrease. For example, the
generating capacity mix in 2000 under the 1979 generation plan is 50 per cent nuclear, 26 per cent
coal, 15 per cent hydroelectric, and 9 per cent oil and gas.) The planned mix of capacities is not the
same as the mix determined purely from the considerations of cost economics, which is about 65 per
cent nuclear (see Chapter 3).
System Expansion Program Reassessment (SEPR) Study
In September 1976, the Board of Directors of Ontario Hydro ordered a complete reassessment of the
corporation’s system expansion programme along with a review of all the factors connected with it. The
assessment was intended to provide a broad framework of information that would facilitate the plan-
ning of the future generation expansion programme. One of the reasons for undertaking SEPR was to
respond to a June 1976 report of the Ontario Legislature’s Select Committee on Ontario Hydro Affairs,
which recommended the adoption of a revised generation programme with a reduced target for addi-
tional generating capacity, after the implementation of load-management and conservation
programmes.’
The purpose of the SEPR study was to estimate the socio-economic effects on the Ontario community of
various hypothetical generation expansion programmes for the period 1978-97 and to examine the
relationships between the growing demand for electricity, the mix of nuclear and coal-fired generat-
ing capacity, the size of the generating units, the reliability of electricity supply, and the cost, availabil-
ity, and security of fuel supplies, on the one hand, and financing requirements and capital availability,
the cost of electric power, and socio-economic and environmental conditions, on the other hand.
The SEPR study represents an important step in the evolution of Ontario Hydro’s planning process. It
is basically an evaluation of many possible generation expansion programmes as responses to two
possible demand growth rates. The study is one of several possible methodological approaches to the
issues studied. Every approach has certain strengths and weaknesses. The major strength of the SEPR
work is its detailed analysis of the reliability, cost, and broad economic impact of the generation-load
The Planning of the Electric Power System in Ontario 81
scenarios studied. Its principal weakness is the fact that the generation expansion programme to be
evaluated is specified at the outset and is not modified even if it becomes clear that it would have an
adverse economic impact on society. Thus, the methodology is a good one for evaluating a set of genera-
tion programmes but not for selecting a generation expansion programme based on given system
design criteria.
An assumption fundamental to the study’s evaluation framework is that a specific predetermined
generation expansion programme will be followed throughout the study period no matter how low the
system reliability falls or how large the costs of power become. Thus, none of the adaptive possibilities of
the programmes, as normally implemented, is incorporated in the study. Moreover, except for a new
and important approach to reliability, no uncertainty was incorporated into any element of the results.
It should be emphasized that this type of study is an important improvement in the methodology used
earlier and should be regarded asa first step along an important new path.®
The study considers variations in the generation expansion programme for only two load-growth rates
— 6.4 per cent and 5.5 per cent average annual growth. All variations assume the completion of the
committed generation expansion programme, including the Darlington Generating Station. The vari-
ations in the uncommitted generation programme consist of three levels of mix of coal-fired and
nuclear generation — high nuclear (2 nuclear to 1 coal), low nuclear (1 nuclear to 2 coal), and no nuclear
(i.e., all coal); two levels of base-load generating unit size — 850 MW and 1,200 MW; and five levels of
target generation reserve — 15, 20, 25, 30, and 35 per cent. The study did not consider variations in
transmission and distribution expansion, the location or dispersion of generation, or generating units
of smaller sizes. Since the study was initiated late in 1976 (released in February 1979), the load fore-
casts have dropped well below the lowest growth rate considered in the project. Thus, the methodology
and general nature of the results are the only aspects that are still relevant.
The methodology consists of two stages. First, judgement is used to select a set of generation expansion
programmes in accordance with different assumptions about the external environment. The genera-
tion programmes are characterized by a constant reserve margin, a constant mix of coal and nuclear
units, and a given size of base-load units over a 20-year period (from 1978 to 1997). The growth of
demand is the only major external condition that varies. This choice of expansion plans determines the
results of the ‘what if” type of question. For example, the choice of 850 MW coal-fired units as the
means of varying the reserve margin determines the incremental cost of reserve (that is, the slope of
the “expenditure” line in Figure 4.5 in Chapter 4). Moreover, although this type of study computes the
cost of power for different generation programmes, it does not allow that cost to affect the growth of
demand and thus the associated generation programme. Similarly, this approach does not allow the
reserve margin associated with an expansion programme to be increased, even if the reliability is
decreasing.
The second aspect of the methodology is that, for each expansion programme, various system perform-
ance characteristics such as cost, reliability, and socio-economic impact are computed for each state of
the external environment (in this case the two levels of demand growth). The generation scenarios
were analysed in three phases. In Phase I, fuel requirements, long-run costs, revenue requirements
(cost of power), and borrowing requirements were computed. In Phase II the study focused on the
question of reliability. (This is described in detail in Chapter 4.) In Phase III, various judgemental
modifications to an econometric model were used to estimate for each load-growth rate the relative
effects of each generation programme on major economic indicators such as economic growth, inflation,
employment, balance of payments, and the Canadian dollar exchange rate. In addition, Phase III
estimated the relative manpower requirements, the effects of a programme change on supply and
export industries, and the effect of higher electricity costs faced by some industries.
The results fall into three categories: the effect of the rate of growth of demand, the effect of the mix of
generating types, and assessment of the appropriate reliability standards. For example, if the demand
growth rate is less than 5.5 per cent (as Ontario Hydro’s 1979 load forecast suggests), no critical fuel
supply or financing problems are envisaged. The results also indicate that if the load growth is sufficient
to warrant new installations, a programme with about *% nuclear capacity will give both the lowest long-
run costs and the lowest cost of power. It is important to note that the present value of economic costs
becomes less and less sensitive to the mix chosen for the expansion programme as the rate of growth of
demand drops. For example, the extra economic costs of a no-nuclear programme over a high-nuclear
programme (both with a 25 per cent reserve margin) with a 5.5 per cent annual growth of demand are
82 The Electric Power System
less than half the extra economic costs with a 6.4 per cent annual growth. Note also that while a higher-
nuclear programme is somewhat lower in economic cost than the other programmes, it requires more
capital and so may result in a somewhat higher retail cost of power until the growing fuel cost savings
outweigh the increased capital requirements. The results of the econometric model suggest that a
reduction in the role of nuclear plants in the generation mix leads, for the most part, to lower invest-
ment in Canada and in Ontario. The new approach to balancing the costs and benefits of reliability
indicates that the traditional LOLP calculation gives a reserve margin level (about 30 per cent) thatisa
few percentage points higher than the new estimates (see Chapter 4).
Phase III of SEPR also analysed the economic potential of industrial co-generation in Ontario. The
results suggest that if the system were expanded at the lowest reserve margin considered (that is, 15
per cent), the economically and technically justifiable co-generation capacity in 1985 would be about
1,100 MW. Athigher system reserve levels, the economic potential would drop to 600 MW.’
The main thrust of innovation in the SEPR study was its attempt to be much more inclusive than
methodologies used previously. There are two main new features. The first is the treatment of reliabil-
ity in terms of balancing the benefits to the customer of a given reliability level with the costs imposed
on the customer by the utility for supplying that level of reliability (see Chapter 4). The second is the
attempt to estimate, for each demand growth rate, the indirect economic effects arising from the
generation expansion programme. The focus of this work was the development of econometric models
to reflect Ontario Hydro’s system expansion and borrowing activities. First, models of the Canadian
and Ontario economies were modified so that capital expenditures, domestic and foreign borrowing,
and the price of electricity could be entered as independent variables, Second, the Ontario model was
extended by adding information from the 1965 Input-Output Table for the province, to measure the
output of Ontario industries. Next, forecasts for Canada and Ontario for the period 1978 to 1997 were
developed for the reference case associated with each load-growth alternative. Then a number of
simulations were made by using different sets of data, reflecting variations in system expansion
programmes. The direction and magnitude of the changes in important economic variables were then
used to provide an indication of the nature of changes in the economic environment.
As Ontario Hydro has stated, SEPR is only one component of Ontario H ydro’s changing planning
process, and it does not address all the factors that must be considered in system design. Some aspects of
SEPR that limit its application to system design have been identified. These aspects will be discussed, in
a review of the possibilities for future related work.
The first point is that the characteristics of the generation programmes studied vis-a-vis each load-
growth projection were fixed at the outset and were not allowed to vary with time. The reliability
implications of such an assumption were discussed in Chapter 4. Concerning the generating mix
implications, while the methodology permits the evaluation of various long-term generating mixes, it
does not, for example, determine the effect on generating mix of changes in various planning factors,
such as load growth, capital availability, and fuel supply.
The second point is that uncertainty is not analysed in the strategic sense. For example, the study
assumes that demand uncertainty affects the reliability of the system, but not the choice of a genera-
tion programme or fuel supply. The portion of the SEPR study that deals with reliability makes the
important point that in the presence of uncertainty, unless losses on one side of the average value are
balanced by gains on the other side, the average outcome may be misleading. The other areas of the
study do not recognize this issue, although much uncertainty is clearly present in the estimates used.
For example, in the lower load-growth case, the study concludes that, while all generation alternatives
are financially viable, there are increased risks of capital availability constraints on the high-nuclear
alternatives. Similarly, the study concludes that while a no-nuclear option after Darlington is viable
under the lower load-growth assumption, there would be increasing coal-supply uncertainty. We
believe that in subsequent studies considerable analytical work should be devoted to the problem of
choosing the best expansion programme in the context of explicit representation of the most important
uncertainties.
The third limiting aspect of the study, as discussed earlier, is that neither changes in the cost of power
nor changes in the reliability levels over time affect the demand for electricity and thus the associated
generation programmes.
The Planning of the Electric Power System in Ontario 88
Ontario Hydro’s Current Generation Expansion Programme
Following its review of the generation expansion programme in 1978 (Program Z) and faced with a
further sharp reduction in the load forecast, Ontario Hydro undertook another review of the genera-
tion programme in 1979.°
In the 1979 load forecast, the average annual rate of load growth to the year 2000 is approximately 4.5
per cent, compared with 5.4 per cent in the 1978 load forecast. The forecast total system January
primary peak in the winter of 2000-01 is 43,031 MW — the peak in 1978-9 was 16,252 MW. Two other
considerations characterize the latest review. It incorporates the conclusions of Ontario Hydro’s Sys-
tem Expansion Program Reassessment (SEPR) study with respect to reliability and generating mix
(see the preceding section). The review also includes, as part of the long-range plan, a further develop-
ment of hydraulic capacity which is a proportion of the approximately 2,000 MW hydraulic develop-
ment programme approved by Ontario Hydro’s Board of Directorsin July 1978.
The long-term generating-mix assumptions are basically the same as those used in the previous gener-
ation forecasts except for the inclusion of about 1,100 MW of peak- and intermediate-load hydraulic
development in the 1990s, as mentioned. For the forecast loads, approximately three-fourths of the
capacity additions beyond Darlington are nuclear.
Load management is included in the programme to reduce the primary peak loads. The load-manage-
ment targets, estimated in a study carried out by Ontario Hydro’s Energy Conservation Division in
July 1978 (see Chapter 8), are used to determine the managed firm peak load, which is the basis of
generation planning. The targets are 500 MW in 1985, 1,300 MW in 1992, and presumably about 2,000
MW in 2000. The managed firm peak load is obtained by deducting these load-management targets, as
well as approximately 500 MW of interruptible loads and 178 MW of the Bruce Heavy Water Plant
electrical load, from the primary peak load.
The standard of reliability used in the 1979 review is lower than Ontario Hydro’s previous practice of a
loss-of-load probability (LOLP) of one day in 10 years. The SEPR study concluded that the “planning
standard for generation reliability can be reduced without undue risk to the quality of service, pro-
vided adequate transmission capacity is available.” Consequently, Ontario Hydro has used a new
reliability criterion of 10 system minutes of unsupplied energy per year, based on the “frequency and
duration of outages” method (see Chapter 4). This includes reliance on emergency support of 500-700
MW from interconnections with the neighbouring systems. In calculating the expected unsupplied
energy, explicit account is taken of an assumed 2.7 per cent load reduction to be achieved by a5 per cent
reduction in supply voltage.
The new criterion results in reserve requirements of approximately 23-25 per cent relative to the
managed firm peak or 15-17 per cent relative to the primary peak. This compares with the 27-30 per
cent reserve margin standard (relative to the primary peak) used previously. Thus, the net effect of load
management, of reliance on interconnections, and of equating costs and benefits of reliability is a
reduction in the generating reserve margin of approximately 12-13 percent.
Another noteworthy change in generation-planning methodology evident from the 1979 review is the
joint planning of Ontario Hydro’s East System and West System. The planning assumes a high-
capacity interconnection between the two systems by the late 1980s or early 1990s. However, the
capacity and the nature (AC or DC) of the interconnection are not discussed.
After reviewing several alternatives, the Ontario Hydro Board of Directors adopted, for the committed
generating stations, the programme shown in Table 7.2. The in-service dates of the Thunder Bay
Generating Station and of Pickering B remain unchanged, i.e., as they were in Program Z. A decision
to stop the construction of the Wesleyville Generating Station and store it until 1990 was taken in
February 1979. Atikokan units 1 and 2 have been postponed by one and four years, respectively. Bruce
units 5 and 6 remain on schedule but units 7 and 8 are postponed by one year. Completion of the first two
units at Darlington has been extended by 18 months and of the last two units by 30 months, from the
original schedule in Program Z. Table 7.3 shows the mix of capacity of the existing and committed
generating resources as well as of the additional uncommitted programme to 2000, under the 1979 load
forecast and two lower-growth scenarios (4 per cent and 8 per cent). For the uncommitted programme,
the specifics of unit size, in-service dates, and sites are not discussed. Ontario Hydro’s position is that
these ‘are the subject of detailed study before any recommendation is made for the commitment of
new generating capacity”. As was the case with the earlier long-range generation forecasts (Table 7. 1),
the share of nuclear in the total system capacity for the 1979 forecast load is about 50 per cent by the end
84 The Electric Power System
Fig. 7.3: p. 94
Fig. 7.4: p. 95
of this century. Figures 7.3 and 7.4 show the annual fuel consumption and contracted supply for
uranium and fossil fuels, respectively. Also shown are estimates of fuel consumption under a 3 per cent
load-growth rate, which will be discussed in the next section. The fuel consumption estimates are based
on the fact that because of their lower fuel or operating costs, the available nuclear and hydraulic
resources will be utilized to the fullest extent possible before fossil fuels are used.
Table 7.2 Ontario Hydro’s Committed Generation Programme
In-service date
Station Size (MW) Fuel type 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
Thunder Bay 2 x 149 lignite l 1
Pickering-B 4 X 516 nuclear - 1 2 l
Bruce-B 4 xX 756 nuclear - - - 1 1 - 1 1
Atikokan 2 X 206 lignite - - - - ] - - - ]
Darlington 4X 881 nuclear - - - - - - - 1 1 l 1
Wesleyville 2X 541 oil - _ = = eS = bs
Source: RCEPP.
Table 7.3 Ontario Hydro’s Generation Programme to the Year 2000
Load growth projection
Committed (MW) Uncommitted (MW)
1979 load
forecast @ 4% 3%
Nuclear 13,860 11,450 6,800 0
Fossil® 14,855 2,750 0 0
Hydro 6,515 1,100 1,100 600
Total 35,230 15,300 7,900 600
Total generating capacity in 2000 50,530 43,130 35,830
Notes:
a) Approximately 4.5 per cent.
b) Uncommitted fossil is all coal.
Source: RCEPP.
Implications of Low Load Growth
In its ‘1979 Review of Generation Expansion Program”, Ontario Hydro raised concerns about the
implications of a load-growth rate lower than their 1979 load forecast. These concerns relate to the
cutbacks in the delivery of fossil fuels in the 1980s and the likely impact on the nuclear supply industry
of delaying uncommitted nuclear capacity. These concerns will be discussed here under the assump-
tions of a 8 per cent annual load-growth rate. If Hydro’s primary peak load were to grow at an average
annual rate of 3 per cent, the primary peak in the winter of 2000-01 would be 31,140 MW. As shown in
Table 7.3, the total dependable peak capacity of Ontario Hydro’s existing and committed programme is
35,230 MW, and at a8 per cent load-growth rate the uncommitted programme calls for no new fossil or
nuclear capacity and only 600 M W of hydraulic development to the year 2000.
Ontario Hydro’s concerns relating to the cutbacks in the delivery of fossil fuels arise from the growing
disparity between the contracted supply levels of fossil fuels and their projected utilization under a 3
per cent load forecast assumption:!”
A reduction to a low load forecast of 3 per cent would result in severe cutbacks in delivery of U.S.
bituminous coals in the short term along with reductions in Canadian coal deliveries in the years 1982
to 1989. These cutbacks in deliveries over an extended period would result in a substantial increase in
the unit cost of fuel through penalties for non-delivery and anticipated increases in unit coal prices.
In the extreme scenarios with unmodified generation programs and with 3 per cent load growth,
fossil fuel requirements decline to virtually zero in the late 1980s with consequent severe contract
penalties and impairment of generation diversity.
Ontario Hydro’s concerns are well founded, but the problem of cutbacks in the U.S. coal deliveries will
exist until the mid 1980s even with the current load forecast of approximately 4.5 per cent (Figure 7.4).
Of the approximately 10 million tonnes of fossil fuels that will be required annually during the first half
of the decade, the share of residual oil, natural gas, and western Canadian coal is about 3.6 million
nnn. —
The Planning of the Electric Power System in Ontario 85
tonnes. The remaining 6.4 million tonnes, to be supplied from the U.S. markets, is only 65 per cent of the
contracted U.S. supplies and thus represents a 35 per cent cutback.
With aload-growth projection of 3 per cent, the fossil fuel utilization will indeed decline to virtually zero
by 1990 if the committed programme is not modified in response to the lower loads (Figure 7.4). Such a
scenario appears to be unrealistic. If the load were forecast to grow at 3 per cent, some committed
capacity would be deferred — most likely Darlington and probably Bruce units 7 and 8. Even if units 7
and 8 are completed on the current schedule (1986-7), the availability of the full 6,000 MW output of the
Bruce complex is in doubt due to the uncertainty associated with the second 500 kV line out of Bruce.
This uncertainty was noted by Ontario Hydro in its 1979 review:!!
It is foreseen that a lengthy public participation and review process will be required to establish the
need, the plan, and the route and site location for a second 500 kV line from Bruce. The earliest in-
service date for this line is estimated to be late 1986. However, it is expected that there will be
considerable opposition to this line from segments of the public in southwestern Ontario and the in-
service date could be significantly later.
The Commission, which had been asked to investigate the need for additional bulk power facilities in
southwestern Ontario, recommended in its report to the Minister of Energy in June 1979 that:!”
Because we foresee serious social as well as environmental and economic problems associated with the
possible construction of a second 500 kV line from Bruce following any route that crosses the prime
foodlands of Ontario, all other alternatives...even if there are apparent economic penalties, should
be explored fully before further consideration is given to such a proposal. :
On the issue of deferring units 7 and 8 of Bruce B in case the second 500 k V line is not in service by 1986,
the Commission observed: “There is probably little advantage in completing them before the second
500 kV line is available.” !%
If load growth is an important consideration in the determination of the timing of the second 500 kV
line out of Bruce, a 3 per cent load forecast will tend to delay it.
We consider two scenarios involving deferrals in the committed programme (Figure 7.4). The one with
the lower fossil-fuel requirement assumes the deferral of Darlington by three years from its current
schedule (1987-90), whereas the one with the higher requirement assumes the deferral of Bruce units 7
and 8 by three years and Darlington by six years. Even in the second case, it will be possible to maintain
a 25 per cent generating reserve margin until the late 1990s. As may be seen from Figure 7.4, defer-
ring Darlington by three years increases the fossil-fuel consumption significantly in the late 1980s and
the early 1990s, and this consumption exceeds the contracted amount in 1989. By deferring both Bruce
and Darlington, the consumption exceeds the contract levels in 1987 and after that it averages the level
implied by the 1979 load forecast. After the completion of the committed programme, the fossil-fuel
requirements under the 3 per cent scenarios increase rapidly, and exceed the 1979 load forecast levels
by the late 1990s.
Our overall conclusion based on this analysis is that a reduction in load forecast to 3 per cent will not
increase the severity of impact on coal supplies to the extent predicted by Ontario Hydro. As far as
surpluses in the early 1980s are concerned, Hydro’s efforts to export excess power and energy are
encouraging. The latest estimate of export sales in 1979 is 12 TW-h, up from 10.4 TW-h in 1978 and 8.4
TW-hin 1977. The sales represent approximately 4 million tonnes of U.S. bituminous coal.
Figure 7.3 shows the estimated annual uranium consumption corresponding to the three scenarios
under the 3 per cent load-growth assumption. It is evident that considerable uranium oversupply will
develop under all the 3 per cent growth rate scenarios if the contracts are not changed in response to
lower demands. Due to the nature of the uranium supply contracts and the relative ease of storage,
shortages appear to present much greater risks than surpluses.
Concerning the nuclear supply industry, Ontario Hydro has expressed concern that any generation
programme corresponding to the “1979 (load] forecast [will] have very serious consequences for Ontario
Hydro suppliers of equipment and services,” and that “should the 1979 load growth projections be
optimistic and further delays of the nuclear programme be required, there is a real risk of losing
valuable and necessary technology and expertise”.!4 The Commission presented its views on this issue,
which is outside the scope of this volume, in its Interim Report on Nuclear Power in Ontario and dis-
cusses it further in Volume 1 of this Report.
SSS eee
86 The Electric Power System
Planning Bulk Power Transmission
The process of planning electric power for Ontario has been treated up to this point in global terms, that
is, as the matching of power generation to load growth. The question of interconnecting generation
with load remains to be considered. In order to be supplied properly, new loads must be connected
effectively to the power network and through it to new generating facilities. It is to be expected that the
existing power network may be unable to transport the increased amount of energy and that the
network will therefore have to be expanded. Alternative expansion schemes must be considered and
the best one selected. Expansion of the part of a network that interconnects major load centres with
major generating sites is the concern of what is called bulk power transmission planning.
First, the general nature of bulk power transmission planning will be considered briefly. Then the
planning process pursued by Ontario Hydro will be examined, and, finally, there will bea brief look into
the future of bulk power transmission in Ontario. In the last portion of the discussion, issues related to
the topic will be discussed and an attempt made to identify concepts that are likely to be useful in the
future planning of the transmission network.
General Approach
The bulk power network in Ontario is an extensive one, and it connects many load centres with many
generating stations. The gradual growth in load as well as the creation of new load centres must be
matched by the addition of new generating units or the development of new generating sites, and the
capability of the bulk transmission network to accommodate the added flow of power must also be
considered.
In the planning of an expansion of the network, several often competing requirements must be satis-
fied. First, the transmission system must be highly reliable. The added load and the added generation
must be so integrated that the reliability of the system does not deteriorate. Second, the expanded
network must have acceptable security. Third, the selected plan must be as efficient as possible, so that
the transport of power is accomplished with an acceptable energy loss. And, finally, the expanded
network must be cost-effective and its environmental impact must be minimal.
A detailed study of each plan is required, to satisfy the enumerated requirements. Consequently, the
planning process must commence with the identification of suitable alternatives. In most cases at
present, the initial choice is made on the basis of human judgement. The previous experience of the
planner can therefore be a great asset in this phase of the planning process. Once acceptable alterna-
tives have been identified, the evaluation of each alternative may be undertaken by a more structured
procedure, involving use of acomputer.
The evaluation phase involves several stages. Of these, the technical evaluation can be performed
adequately by acomputer. The power system in Ontariois interconnected with those of its southern and
western neighbours, and for that reason the operating criteria of the interconnected systems must be
compatible. Appropriate design criteria for the security and reliability of the network were developed
by the Northeast Power Coordinating Council (NPCC), of which Ontario Hydro isa member. According
to these criteria, the power system of each member must, for instance, be able to maintain stability
when one transmission circuit is out of service (e.g., for maintenance) and, at the sametime, a transmis-
sion line (with either one or two circuits) is lost perhaps because of a storm. Each proposed alternative
for expansion of the transmission network must comply with these requirements. Questions of result-
ing line flows and of possible overloadings are readily answered by computer studies performed for
each specified set of operating situations. After the technical feasibility has been established, the
economics of each proposal is evaluated, and, finally, the impact upon environment is assessed.
Now the stage is set for a meaningful comparison of the alternatives. A considerable amount of human
judgement inevitably enters this task, because comparisons must be made between quantities that
have distinctly different qualities. Yet this is one of the most important stages of the planning process.
Its results establish which of the alternative expansion schemes should be adopted.
Ontario Hydro’s Approach
In June 1976, Ontario Hydro submitted to the Commission a memorandum describing its transmission
planning process.'° When planning the development of its bulk power transmission network, Hydro
divides the process into four main steps:
e determination of the additional facilities required, and their timing
The Planning of the Electric Power System in Ontario 87
e developmentof alternative systems capable of meeting specified requirements
e evaluation of thealternatives
e selection of the alternative to be recommended
Ontario Hydro considers it necessary, to reduce the overall costs and increase the reliability, that the
selected alternative be compatible with its long-range plans.
The first step is taken upon completion of the generation planning process, in which the size and
location of new generation facilities, as well as the timing of their completion, were determined. The
purpose of this step is, therefore, to answer the question, whether existing bulk transmission facilities
are adequate for the purpose or whether additional facilities will be required. Expected loading pat-
terns are developed and the operation of the whole system is tested on the basis of specified reliability
and security criteria.
The experience of Ontario Hydro indicates that it takes at least eight years to build a major transmis-
sion line. The actual construction work may take only two years, and at least two more years are
required before that, for making the detailed design, acquiring the needed right of way, and prepar-
ing for the construction phase. Before any of this can take place, however, at least four years will have
been spent on developing plans for alternate routes, arranging for public participation in the selection
of acceptable alternatives, and obtaining government approval of the final choice.
To arrive at a number of feasible alternatives is the goal of the second step in Ontario Hydro’s bulk
power transmission planning process. There may be a number of ways in which the existing system can
be expanded to satisfy the requirements of the assumed load growth and of the pattern of additional
generation. These alternatives may range from the simple rearrangement of connections in the exist-
ing network to the construction of completely new transmission lines and station facilities. They may
involve increasing the rating of lines, upgrading the power transfer capability of existing rights of
way, installing series or shunt capacitors and reactors, establishing new interconnections with neigh-
bouring utilities, and developing facilities for the rejection of generation or of load.
The planner must exercise a considerable amount of judgement in drawing up a list of acceptable
alternatives. To be considered acceptable, a plan must be technically sound. This can be determined
quite rationally because the technical performance of a system can be quantified. Also, the level of
reliability of a plan that is under consideration must be acceptable. It must be ascertained that the
system would operate under normal and emergency conditions within acceptable thermal limits and
within the prescribed voltage range. Short-circuit limits and stability requirements must also be met.
However, a plan that is technically feasible may have to be rejected because of socio-environmental
considerations. Computer assistance is sought where suitable at this stage. Ontario Hydro has devel-
oped an elaborate computer-aided approach to help the planner evaluate the environmental, demo-
graphic, and socio-economic impacts of a projected transmission route. In this approach, the informa-
tion about the region in question is stored in a computer on the basis of a2km by 2km grid. Each square
is then classified with respect to nine “factors” — human settlement, agricultural production, timber
production, mineral extraction, wildlife game resources, recreation, aquatic communities, terrestrial
communities, and the appearance of the landscape. Each factor is further divided into a number of
“objectives”. For example, the factor “agricultural production” is divided into eight objectives, such as
areas producing fruit, vegetables, and tobacco, areas with a very high concentration of common field
crops, areas with a low concentration of common field crops, class I or II soils officially designated for
future residential, commercial or industrial use, etc. A total of 49 such objectives are then placed in
order of their importance with respect to selecting a route. For example, existing urban and non-urban
areas of human settlement are at the top of the list and must be avoided. With this information base in
the computer, the planner can make rational decisions about the impact of any proposed routing. A
composite map can also be produced, enabling the planner to see how routes of significantly undesirable
impact can be avoided. .
When the planner has identified a plan that has a high probability of satisfying the many varied
requirements for the expansion of the transmission system, a detailed evaluation of that plan is under-
taken, involving the compilation of environmental data, cost data, and technical data. These back-
ground data are put to use in the final step, which is the selection of the “best” alternative. The environ-
mental information includes details about ecological impact, land productivity, and the probable effect
on human environment of each feasible plan; these details are considerably more extensive than those
used in the computer-assisted assessment mentioned above. As for cost data, the effects of capital cost,
Tat RIE orec remem eee Cine ee Ee eee I Oe
88 The Electric Power System
operating expenses, and the cost of transmission losses and maintenance are included. To remain in the
running, a plan must meet certain minimum technical requirements. But every plan has a somewhat
different impact upon such factors as reliability, ease of operation, and ability to adapt to variations in
the economic conditions of the province. Some alternatives may be more adaptable, for instance, to
changes in the load forecast. All these factors are taken into account by Ontario Hydro when it decides
to accept a particular proposal for the expansion of its bulk power transmission systems.
Planning for the Future
The preceding discussion provides some idea of the process of bulk transmission planning, as back-
ground for an examination of some of the important issues in the planning process. Discussion of the
expansion of the bulk transmission facilities in Ontario cannot be limited to a consideration of load
growth and new generation. The planned retirement of outdated generating plants and the replace-
ment of old bulk transmission facilities must also be considered. Additional lines may be needed, not
only because of expected growth in load but also to replace obsolete transmission equipment. Replace-
ment of obsolete generation plants will not affect the extent of required transmission facilities, but it
may indicate a need to change the configuration of the bulk transmission network, should the replace-
ment generation be at a different site.
The strategy for developing bulk power transmission is roughly similar to the principle used for the
installation of a local distribution system. It is easy to see why, in a row of houses in a subdivision, an
electricity supply configuration in the form of a mesh is greatly preferable to a radial system. With a
mesh, the row of houses is supplied from both ends, while in a radial system, it is supplied only at one
end. When a fault occurs along the feeder, in a radial system, service is interrupted to all houses that are
located beyond the fault. Security of supply is, therefore low. When the same row of houses is supplied
from both ends, the occurrence of the fault can only cause an interruption of supply to one house, at the
most. Thus, the security of supply is much greater.
The experience with distribution systems may be applied directly to the bulk transmission system. The
main difference is that security criteria for the transmission system must be much higher than those
for a distribution system, because far more customers are relying on the transmission network. In-
deed, some loads are so critical that they may not be left without supply under any circumstances.
Hospitals, for instance, are in that class; for them, in addition to providing a supply of high security, an
emergency generator, usually with a diesel engine as the prime-mover, is installed on the premises.
To create a bulk power transmission system of very high security for a region, the same mesh concept
that is used for local distribution is applied, on alarger scale. Each load centre, such as a city, is supplied
from at least two generating sites, and the bulk transmission system is so designed that power con-
verges on the city from geographically opposite ends. The generating sites are also connected directly
with one another, so that this part of the bulk power network has the shape of a triangle. Power can be
delivered tothe city from both generating sites during a fault on any side of the triangle.
An example of this approach is provided by the bulk power transmission systems in southwestern
Ontario and eastern Ontario.!® In southwestern Ontario, the planning of the bulk power transmission
system must meet three basic objectives: to supply the growing loads in the region, particularly in the
London, Sarnia, and Windsor areas; to incorporate Bruce B into the bulk power grid; and to strengthen
the transmission system so as to facilitate the mutually beneficial interchange of power with U.S.
utilities. The expected growth of the load in the region will, if realized, require additional transmission
facilities in the 1990s. The simplest way to satisfy the three aforementioned objectives would be to
construct one 500 kV line to London from the Bruce generating complex and another from the Middle-
port Transformer Station near Nanticoke. A triangle of the 500 kV grid would thus be constructed in
southwestern Ontario, the third leg being the 500 k V line from Bruce to Milton and down to Middleport
and Nanticoke. A detailed study of this proposition reveals a possibility of achieving the stated aim
while eliminating the need for part of the right of way that would otherwise be required. The saving
may be accomplished by taking down one of the three parallel 230 kV circuits between Middleport and
London and utilizing its right of way for the new 500 kV line. The possible scenario may be described as
follows.
First, a 500 kV line would be constructed to reach London from the north and supply London with
electricity generated at Bruce. Then the loading on the two 230 kV lines from Middleport to London
would be decreased to the extent that one of their three circuits could be taken out of service for several
years without detriment to the supply. At this stage the single-circuit 230 k V line would be pulled down
The Planning of the Electric Power System in Ontario 89
and a 500 kV circuit erected on its right of way. The timing of the changeover would be extremely
important. Conversion would have to be completed before the load in the London area increased to the
extent that the security of supply would be contingent on the 500 k V circuit from Middleport.
It is emphasized that serious social and environmental concerns are associated with the possible con-
struction of a second 500 kV line from Bruce over prime foodlands. These concerns must be resolved
before such a proposal is adopted. The Commission underlined these concerns in June 1979 in its Report
on the Need for Additional Bulk Power Facilities in Southwestern Ontario.
In eastern Ontario, the immediate and most important consideration in the planning of the bulk power
transmission system is the supply to the Ottawa area. Because of transmission deficiency, automatic
reactive power devices will have to be depended on in the 1980s to maintain voltages in the Ottawa area
at acceptable levels following any failure in the bulk power transmission system. The level of load that
could be supplied will be exceeded in the early 1980s under any annual rate of load growth in excess of 3
per cent. The planning of additional bulk power transmission facilities in eastern Ontario is therefore
a matter of urgency. The Commission recommended the continuation of the planning process, in July
1979 in its Report on the Need for Additional Bulk Power Facilities in Eastern Ontario.
One of the most technically attractive ways of supplying Ottawa by means of the existing generating
resources, using the triangle concept, is to construct a triangle of 500 k V lines with Ottawa, the Lennox
Generating Station near Kingston, and the Saunders Generating Station near Cornwall as the three
vertices. This would ensure the supply to Ottawa from two major generating stations and strengthen
the bulk power grid in eastern Ontario in a way that would facilitate mutually beneficial interchanges
of power with Quebec and New York. As in the case of the projected 500 kV line in southwestern
Ontario, it isemphasized, all other alternatives must be explored fully. .
At the present time, human judgement constitutes a very important component in the decision process
for new bulk power transmission components. This means human participation in the process, which is
desirable, since planning involves the use of innovative ideas and should not be completely delegated to
a computer. However, there are many steps in the development of background arguments for the
decision-making process that can and should rightly be carried out, in our age, by acomputer.
One of the tasks that may be usefully performed by a computer is the testing for security of an
expansion alternative. The process of arriving at this information is usually referred to as the “‘contin-
gency analysis”, and it is at present still an immature area of power system studies. In this approach,
many calculations must be made to determine what operating situations might cause an overloading of
a given transmission system. In the case of Ontario Hydro, such a study may involve hundreds of load-
flow solutions for each plan being considered, and the practice up to now has been to allow human
judgement to intervene at this stage of the planning process. However, the present trend in system
studies is directed towards the development of very fast, although approximate, computer techniques
to establish the effects of changes in the transmission network upon the associated distribution of
power flows. The results are not exact, but they are accurate enough to indicate when a proposal for
network expansion should be rejected. Such computer approaches are expected to become available in
the near future as an aid to the system designer in selecting expansion alternatives that should be
studied in more detail.
Something that is lacking at present in the process of planning bulk power transmission is a systematic
approach to the problem of weighing the technical aspects of a plan against the environmental consid-
erations. At present, responses to questions of the environment tend to be emotional, whether pro or
con. But there would be no need for this if it became possible to make rational comparisons. What is
needed is an approach that would provide a meaningful assessment of the environment impact and
thereby permit a comparison of technically feasible alternatives. Ontario Hydro’s factor map is an
important step in this direction.
Concerning the stop-gap measures proposed by Ontario Hydro!” for upgrading the bulk power trans-
mission network, these often constitute an economically viable and ecologically attractive way to cope
with the expansion of load and generation in Ontario. Nevertheless, care should be taken in each case to
ascertain that the proposed solution is the best one. While effective in the short term, a solution may
turn out to be less than desirable in the long run. The upgrading of the current-carrying capacity of
some of the 115 and 230 kV lines in Ontario is a case in point. Such an approach would help to postpone
the need for the 500 kV network in southwestern Ontario by loading the existing lower voltage circuits
to the fullest. At the same time, such a choice would lessen the flexibility of planning the transmission
90 The Electric Power System
system. In the case of the eastern Ontario system, for example, it would preclude a later conversion of
115kV linesto 230 kV.
Another matter that should be considered in connection with the bulk power transmission in Ontario is
high-voltage direct-current (HVDC) transmission. To date, there has been no need for any HVDC
facilities in the province, but this situation may change, for two reasons. The first has to do with the fact
that HVDC is cheaper than high-voltage alternating current (HVAC) transmission over great dis-
tances (see Chapter 2). Consequently, if there is a need to develop remote generating sites or to inter-
connect systems at great distances, the HVDC alternative becomes attractive. Secondly, im-
provements in solid-state terminal equipment for HVDC systems mean that the cost of HVDC
terminal stations is dropping steadily and the break-even point between the HVDC and HVAC is
moving towards smaller and smaller distances. At present, the break-even distance is about 650-800
km.
There are three possible future uses for HVDC transmission in Ontario. One of these is an interconnec-
tion between Ontario Hydro’s East System and West System and between the West System and Mani-
toba. Manitoba has the potential for further development of its hydraulic generation sites on the
Nelson River. For that reason, Manitoba might become interested in an interconnection with Ontario.
Such an interconnection would also be of benefit to Ontario Hydro, since it would strengthen its bulk
power grid in northwestern Ontario and provide a strong interconnection with a neighbouring prov-
ince. Another possible application of HVDC may be envisaged in connection with the development of
the lignite deposits in the Onakawana basin in northern Ontario. The distance to the closest significant
load — Sudbury — is about 500 km, and it is another 400 km to Toronto. There is little settlement along
the route, which is a desirable situation for point-to-point HVDC transmission. Ontario Hydro is seri-
ously considering the construction of a 1,020 MW lignite-fired station at the Onakawana site. The third
possibility for HVDC application is an asynchronous link with Hydro-Quebec (see Appendix C). On-
tario Hydro is studying such an application with an initial capacity of 1,000 MW, ultimately growing to
3,000 MW. .
The Planning of the Electric Power System in Ontario 91
19)
6/ 8/61
8 L8 98 68 v8 £8 28 18 08
De eel g
WB 44] Ea ee
eee ae
(Cina aie mie
GG ‘A
zp ea,
96 $6 v6 £6 06 16 06 68 8
02
Eee
Ob
09
0L
(wajsAg 4823 OIPAY OL2IUQ) Ag WHY JO aBe\UadIag e Se Ajloedeq JeajoNN pauUe|g {*p “aINBIy
Jao Jaq
ddJ0¥ *89N0S
92 The Electric Power System
7 wels0ld
‘dd30¥ *89N0S
_ isi
mE
ua Jag
(waysAg 182] OPH OLeyUQ) yead UI J9A0 UIBeW adiasay Buljeseuay pauue|q Z°/ “@sNBly
ing of the Electric Power System in Ontario 93
The Plann
“OUPAH OLeJUQ pue J47)y :SaanoS
poh
6661 18 0861
| =| ie oa a a peme oge
Sieah pakejap uoBuiyieg ‘sueak ¢ padejap g pue / adMg ‘yAMOsB peo] %¢ —UOlJdwNsUO| Sie ee Neame roe
ee
00S°€
(ospAH O1eJUQ) puewag pue Ajddng wniues] gy “euNdI4
006
000'1
00S'T
pasueyoun awwessosd pajiwwod
‘yymois peo; %¢ —uoljdwnsuo)
WniuedN 40 Swedes
000'2
00S'2
000'€
94 The Electric Power System
aac
6661
PO ae VS
sueak ¢ pacejap uoyBuipeg ‘yymoss peo) %¢ — uoljdwnsuo) HEA
sueak g paXejap u0ysui eq
‘syeak ¢ pakejap g pue / song
‘yyMol3 peo) %F — uoljdwnsuo7)
Ee
Hone 189910} peo| 6/6] — uolydwnsuo7
OUPAH OEIC PUR Jq}JY “SBOINOS
0861
SY PaBUeYOUN aWWeJBOId Pay}iWWOD ‘yMold Peo] %¢ —Uo!dwnsuog
(OJpAH O1e\UQ) puewag pue Ajddng jan4 |Issoj p’y “eunBl4
01
QUUO} Jad (14 UOl||IW G°gz Je JUa}eAINba |209 JO SAUUO} JO SUOI]|IW
él
val
stemin Ontario 95
ver Sy
The Planning of the Electric Pov
gw SUS, ae. ar: -
CS Mle 0 meng aly sory VS al quater NR age 9
: a ae 70. wt iy >
ee 7 “| - a > 7 ae _ oe ion et we .
CHAPTER EIGHT
The Impact of Alternate Technologies
The planning of Ontario’s electric power system will be influenced over the next two decades by the rise
of some so-called unconventional energy technologies, as well as by an increased use of certain conven-
tional technologies. This chapter deals with those new technologies that are likely to have the greatest
direct affect on the bulk power supply system over the period 1981-2000. These technologies are load
management, electric energy storage, co-generation, and generation from biomass and refuse-de-
rived fuels. Strictly speaking, these technologies are neither new nor unconventional. In one form or
another, they have been used in Ontario and elsewhere in the world for many years. What is significant
about them is the recognition of their growing potential. The likely impact of these technologies will be
described, with reference to system load patterns, the mix of conventional generating resources, and
system reliability.
Other embryonic systems for electric power generation, such as solar photovoltaic, solar thermal, wind
turbines, magnetohydrodynamics (MHD), geothermal, and fusion, are not expected to have any sig-
nificant impact on Ontario’s electric power system before the end of this century. Some remote power
systems, such as wind-diesel and small-scale hydroelectric systems, may be suitable for isolated commu-
nities in northern Ontario that are not connected to an electric power system. Solar space heating and
water heating are expected to make increasingly important contributions to Ontario’s total energy
needs, and the Ministry of Energy aims to have 1.5 per cent of total primary energy coming from this
source by the year 2000.' Since solar heating does not directly affect the electric power system, it will not
be discussed. Volume 4 of this Report provides a comprehensive description of both conventional and
emergent energy technologies.
Load Management
“Load management” refers to combined efforts by an electricity utility and its customers to shift the
consumption of electricity from traditional peak hours to periods of lower demand, in order to reduce
the utility’s overall peak demand and consequently increase the load factor. The benefits of load man-
agement are: a reduced need for new generating capacity; an increased utilization of the most efficient
generating units and of transmission and distribution facilities; opportunities for lower customer
rates; and increased business for manufacturers of electrical and electronic equipment through the
sale of heat storage devices, time-of-day meters, and sophisticated electronic communications and
control systems for controlling customer loads. Load management may be summed up, then, as the
shaping of customer load profiles to achieve the most efficient utilization of asystem’s supply capacity.
It does not necessarily imply the conservation of electric energy. Nor does it mean supply management
by such measures as interconnections with other systems or energy storage by the utility, although
these have the same objective.
The methods of load management may be broadly classified into three categories: pricing schemes and
incentives; load control; and load reduction through voltage reduction, appeals to customers, load-
shedding, and load-rationing. Pricing schemes such as time-of-use rates may play a significant role in
the long run in shifting peak loads to off-peak hours. The marginal cost of supplying an additional
kilowatt hour is, generally, higher during periods of high demand than during periods of low demand,
because generating capacity is loaded according to merit order. Thus, with a pricing scheme based on
this concept, there will be incentives to the customers to move some of their consumption from peak
hours to off-peak hours. Time-of-use pricing schemes have received considerable attention among
North American utilities over the last decade. One pioneer is the Long Island Lighting Company
(LILCO), where time-of-use rates for commercial customers with a demand higher than 750 kW have
been in effect since February 1977. LILCO is planning to extend the scheme to residential customers
who consume more than 45,000 kW-h per year. Ontario Hydro is studying time-of-use pricing for its
system and this study is being reviewed by the Ontario Energy Board.”
A form of load management that uses rate incentives has been practised in Ontario for many years in
the interruptible service that is offered by Ontario Hydro to its large industrial users. This service,
which is offered at reduced rates, may be interrupted during a system emergency or for reasons of
economy. The total interruptible loadin Ontario Hydro’s system is about 750 MW.
The Impact of Alternate Technologies 97
“Load control” refers to direct control of customer loads by the utility, for example, the control of
residential water-heaters. It has been used by many municipal utilities in Ontario to a limited extent.
Load control, like interruptible service, can be exercised for economy as well as in an emergency. The
control of residential water-heaters is an example of economic load control; decorative lighting and
residential clothes-dryers are suitable applications for emergency load control. An indication of the
potential for load control may be gained from the fact that about 10 per cent of the electricity used in
1974 in Ontario was consumed by residential and commercial hot-water heaters.°
The third approach to load management is used only in the event of an extreme emergency. The order
in which the measures are imposed is voltage reduction, appeals to customer, load-shedding, and load-
rationing. Voltage reduction and customer appeals were used by Ontario Hydro during the winter of
1976-7. Load-shedding is practised routinely in many developing countries. Load-rationing was used
in the United Kingdom in 1976 and in California in 1974 and 1976.
Experience in Ontario indicates that there are essentially no negative customer impacts as a result of
voltage reduction, provided that advance notice is given. It also indicates that a 5 per cent voltage
reduction can lower the peak demand by about 2.7 per cent. This, then, appears to be an effective load-
management technique to help maintain system reliability. Ontario Hydro has started to take account
of this potential when calculating the reliability of its generation expansion programmes.‘ The experi-
ence during the winter of 1976-7 also proved that appeals to consumers at atime of dire emergency are
effective — a 250 MW load reduction was effected by appealing to consumers through the mass media.
Load reduction can also be achieved through appeals to industry, the results of which are easier to
predict, primarily because of the close contacts between Ontario Hydro and its large industrial custom-
ers. This potential is estimated to be between 400 MW and 600 MW. The total potential load reduction
through appeals to consumers amounts to about 5 per cent of the system demand. This potential can
only be realized on a short-time emergency basis. Much less would be available on an extended or
regular basis. An example of a load-shedding scheme is the one installed by Ontario Hydro in the
Ottawa area to shed up to 300 MW, to avoid a voltage collapse, especially during the period of upgrading
work on critical transmission circuits.” However, Ontario Hydro does not favour load-shedding or load-
rationing as part of its normal system plan.
When considering load management, the distinction between the daily load factor and the annual load
factor must be borne in mind (see Chapter 2). The annual load factor encompasses the weekly and
seasonal as well as the daily variation in load. Load management is generally concerned with improv-
ing the daily load factors, although in doing so the annual load factor is also improved because the
annual peak is thereby reduced. But load management does not aim at shifting loads from one season to
another, because this is not possible at present and it may not be a desired option. Utilities such as
Ontario Hydro take advantage of seasonal variations in the load by carrying out the necessary mainte-
nance work on their generating equipment during the months of lower demand, thus avoiding the
need for additional generating capacity for this purpose. If there is still an excess capacity in the
months of lower demand, it can be utilized through some measure of supply management, such as a
seasonal diversity exchange agreement with another system, whose annual peak occurs in a different
season. Two systems that might benefit from such an agreement are Ontario Hydro, which is winter
peaking, and the New York Power Pool, which issummer peaking.
The cost of implementing load management in Ontario is not precisely known. Investment in load-
management hardware will have to be undertaken by all the parties involved — Ontario Hydro, munici-
pal utilities, and the consumers. A study done for Ontario Hydro to provide a preliminary indication of
the cost of load control in selected industries showed that in some industries only capital costs or only
operating costs will be incurred, while in others both capital and operating costs will be incurred.®. The
estimates range from zero to $830/k W reduced, in capital costs, and zero to $51/kW reduced, in annual
operating costs. Hydro’s estimates of the cost of load management to residential customers show that
large water-heaters (with a capacity to store hot water for longer periods) will cost $100 to $150 more
than the smaller, conventional units.”. The cost to the utility of various load-control systems (AM radio,
telephone, distribution automation) is estimated to be between $250 and $450 per customer location for
prototypes and less than $100 for mass-produced equipment. These estimates are very preliminary and
Ontario Hydro expects to obtain more accurate estimates through field trials carried out in co-operation
with various municipal utilities.
In the Ontario Hydro system, about 6,400 MW of capacity is hydraulic, more than half of which is
suitable for peaking and intermediate loads. Since hydraulic generation has the lowest operating costs,
SURE ee
98 The Electric Power System
and since hydraulic units have excellent load-following capability, there is no incentive to manage the
part of the peak load that is supplied by existing hydraulic generation. In such a case, the maximum
theoretical potential for load management may be estimated by completely levelling off the load on
thermal capacity. Ontario Hydro estimated this potential in its report on “The Role for Load Manage-
ment in Ontario” to be about 3,700 MW in the year 2000, based on its 1978 load forecast and assuming
no further development of hydraulic capacity or storage schemes.® The estimates are derived from the
operation of thermal and hydraulic generating capacity on a December peak day in the East System.
Ontario Hydro chose 50 per cent of the theoretical potential, that is 1,850 MW, as the 1978 target for
load management by 2000, recognizing that major work to study the impact on the system of load
management and other alternatives was still to be completed. The 50 per cent criterion also applies to
the period 1985-99. This results in a target of 700 MW in 1987 and 1,400 MW in 1995.
There have been some developments since the publication of Hydro’s load-management report. They
may affect the load-management programme significantly and therefore deserve comment.
Ontario Hydro’s 1979 load forecast is lower than its 1978 load forecast. Also, in recent years, there has
been ashift in the winter peak from December to January, and the 1979 load forecast assumes that the
January peak is about 3 per cent higher than the December peak. The Commission’s estimates, based
on the 1979 load forecast of January peak for the East System, indicate that the theoretical potential
for load management could drop from 3,700 M W to 8,000 MW in the year 2000, assuming the same daily
load factor. If the load grows only at an annual rate of, say, 3.5 per cent, this potential could be reduced
to 2,200 MW.
Ontario Hydro is also developing a programme for the future expansion of peaking and intermediate-
load hydraulic generation. In its “1979 Review of Generation Expansion Program”, Hydro indicates
that about 1,100 MW of this capacity is planned for the 1990s.° The cost of this development is estimated
at $700/k W (1977 dollars). Since peaking hydraulic generation is an alternative to load management,
the cost of installing such capacity must be weighed against the cost of load management to achieve the
same objective. It is hoped that Hydro will acquire sufficient experience in load management in the
1980s to be able to assess the cost and potential for load management in the 1990s. This will create a
better framework for making a decision on the peaking hydraulic potential. Development of 1,100 MW
of this potential could decrease the maximum potential for load management by about 500 MW.
A third consideration is the development of energy storage (see the next section of this chapter).
Studies by Ontario Hydro have indicated that underground pumped hydraulic storage can be an eco-
nomic alternative, when compared with new coal-fired generation, for supplying the peaking require-
ments. The future development of pumped storage will be determined not only by the growth in load
but also by the economic competitiveness of pumped storage compared with load management. It is not
expected that pumped storage will be needed until the mid 1990s. By then, more experience will have
been gained in load management, and this will facilitate the choice of an optimum mix between the two
alternatives.
The long-term impact of load management on the generating mix is to reduce the need for peaking
capacity and increase the share of high-capacity-factor generation such as nuclear and base-load
hydraulic. For example, in the Ontario Hydro system, if the forecast growth in load is such that addi-
tional capacity will have to be added for peaking, intermediate-load, and base-load application, then
load management will clearly reduce the need for peaking and intermediate load capacity. This is seen
in Ontario Hydro’s 1979 expansion plan, in which the ratio of base-load (nuclear) to peaking- and
intermediate-load (coal and hydraulic) capacity additions in the 1990s is approximately 3:1, compared
with an earlier ratio of 2:1 without load management (see Table 7.3 in Chapter 7).
However, in the short to intermediate term, load management may delay the need for base-load capac-
ity. For example, if the load growth were lower, say 3.5 per cent per year, the existing and committed
peaking and intermediate load resources would be sufficient to maintain an economic mix to the year
2000, and the additional requirements would be primarily for base-load capacity. Now, if the peaks
could be reduced by load management, the utilization of the existing and committed fossil-fuelled
resources could be increased, thus delaying the need for some base-load capacity. The higher f uel cost of
fossil generation may, however, offset much of the capital cost savings.
Load management reduces peaks, but, if the system capacity is reduced in proportion, a reduction in
system reliability must be expected. The reason for this is that, as the daily load factor increases, the
daily load curve flattens, and there will be a greater exposure of system load to equipment outages. This
The Impact of Alternate Technologies 99
is not to suggest that some reduction in capacity is not possible but to point out that the percentage
reserve margin (with respect to the managed peak) will have to be increased if the same level of
reliability is to be maintained. On the other hand, load management, and especially load control, re-
duces the uncertainty with respect to the daily load profile, and this may benefit the system planner by
reducing the risk of having too much or toolittle generating capacity.
Ontario Hydro’s load management initiatives are worthy of support. Hydro is relatively inexperienced
in load management and its participation with the municipal utilities over the next few years will
provide valuable experience in assessing more accurately the cost and public acceptance of load man-
agement. To determine the long-term potential of load management, other alternatives such as peak-
ing and pumped hydraulic resources must be considered. Load management also offers some challeng-
ing opportunities for the province’s electronics industry to design one-way and two-way
communication systems for the monitoring and control of loads at the customers’ premises.
Electric Energy Storage
Due to the nature of the daily demand for electricity it may be desirable to store the surplus output of
low-cost base-load stations during the hours of low demand and use the stored energy during the hours
of high demand. The economic attractiveness of a storage scheme depends upon its capital cost, reliabil-
ity, storage capacity, and operating efficiency, on the availability of low-cost surplus energy, and on the
cost of alternative ways of supplying the daily peaks. A storage system may also increase the operating
flexibility of the total generating system by reducing the need for load-following by large base-load
units such as nuclear units.
No practical utility-scale techniques are available at present for the storage and recovery of electric
energy directly. However, the storage of electricity in the form of the potential energy of water has
been practised for several years. This is generally known as hydraulic pumped storage. By means of
reversible motor-generator and pump-turbine units, water is pumped up from a lower reservoir to a
higher reservoir during low-demand periods and this water is then used to generate electricity when
needed. Ontario Hydro has one such plant at the Sir Adam Beck Generating Station at Niagara Falls,
with a peak capability of about 100 MW. Although most hydraulic pumped storage systems around the
world are located above the ground, underground pumped storage holds considerable potential. In such
a scheme, the lower reservoir is excavated below the ground level, and a large nearby body of water
serves as the upper reservoir. The pumping-generating station is located underground, and only the
control building, transformers, and switchyards are visible on the surface. Thus, from an environmen-
tal point of view, underground pumped storage is preferable to an above-ground scheme.
Several storage schemes have been studied by Ontario Hydro. Significant among the above-ground
pumped storage sites studied since 1965 are Delphi Point (2,000 MW) and Matabitchuan (400 MW). A
preliminary study was carried out in 1974 to evaluate other storage schemes and identify those deserv-
ing further study for development in this century.!° A major thrust for these studies was the need to
reduce load-following by CANDU nuclear units that had been planned at that time for the 1980s and
the 1990s. The 1974 study indicated that potential candidates for more detailed investigation were
both above-ground and underground hydraulic pumped storage, compressed-air storage, feedwater
storage, steam storage, and battery storage (see Volume 4 of this Report for a detailed description of
various storage schemes).
In order to assess the technical and environmental feasibility of the underground pumped storage
concept and its cost, Ontario Hydro authorized Acres Consulting Services Limited to undertake a study
whose results were published in January 1976.'! The report concluded that “underground pumped
storage, utilizing Lake Ontario as an upper reservoir, is a technically feasible means of providing
energy storage for the Ontario Hydro system” and that “environmental considerations associated
with construction activities will play an important part in the siting of the plant” at any location along
the north shore of Lake Ontario.!* The report further suggested that the “economic feasibility of
underground pumped storage must be determined by evaluating the concept within the Ontario Hydro
system using the parameters and costs set out in this study.”!®
Ontario Hydro has, since late 1978, been examining the applicability to its system of the potential
storage schemes mentioned earlier.'4 From these studies, it appears that underground pumped storage
is the most economic of the various storage alternatives for consideration in this century and that the
Hydro system could accept about 2,000 M W of pumped storage by the late 1990s on the basis of the 1979
100 The Electric Power System
load forecast. The ranking of other storage schemes from the most economic to the least economic is as
follows: above-ground pumped storage (Delphi Point and Matabitchuan), compressed-air storage, nu-
clear feedwater storage, lead-acid battery storage, and nuclear-steam storage.
Concerning the role of pumped storage as a peaking plant, Ontario Hydro’s studies indicate that
pumped storage would be economically more attractive than new coal-fired capacity, even if no surplus
nuclear energy was available for pumping, that is, if pumping was done by power supplied by coal-fired
units. However, the degree to which pumped storage could replace new coal-fired capacity will depend
on the required annual capacity factor. Any new coal-fired capacity is most likely to be operated to
supply the intermediate load, that is, annual capacity factors of 10-55 per cent. The maximum annual
capacity factor of which the pumped storage schemes are capable, assuming a daily generating period
of eight hours, is approximately 33 per cent. This, coupled with the existence of considerable amounts of
fossil-fuelled capacity and an upper limit on the availability of surplus nuclear generation, puts an
upper limit on the acceptability of pumped storage in the bulk power system. It should also be borne in
mind when comparing pumped storage with other generation alternatives that it is not a net producer
of energy and thus does not increase the reliability of the system’s energy supply. In fact, pumped
storage increases the system’s primary energy requirements because of the inefficency of the pump-
ing-generating cycle, which is approximately 70 per cent.
This disadvantage of pumped storage is offset to some extent by the fact that hydraulic units have much
lower forced outage rates than fossil-fuelled units, which suggests that, during peak periods, pumped
storage units havea higher load-carrying capability.
In the light of the preceding discussion, Ontario Hydro’s conclusion that “for early inclusion in Ontario
Hydro’s generation program, the most economic large-scale energy storage alternative is pumped
storage, either above-ground or underground, and...no more than a watching brief should be kept on
alternative storage technologies” appears valid.’° However, the potential for use of storage schemes in
the system is uncertain, to the end of the century. The single most important factor will be the growth
in demand. If the forecast growth rate falls below 4 per cent per annum, Ontario Hydro’s estimated
potential of 2,000 MW will be reduced substantially, and, if the load grows at only 3 per cent per annum,
there may not be any need for storage until the next century. This is particularly relevant in the light of
Hydro’s plans to make greater use of the province’s remaining potential for peaking- and intermedi-
ate-load hydraulic generation (see Chapter 7). The economic justification for storage will also be influ-
enced by the cost and success of Hydro’s load-management programme, which has the same objective
as storage.
Co-Generation
Co-generation refers to the utilization of the steam as well as the electricity at a thermal generating
site. It is of potential importance where there is a demand for both steam and electricity, for example,
for industrial processes, and for universities, hospitals, and commercial buildings. Steam can be used as
process-steam or for space heating and water heating. The significance of co-generation as a supply
option for the future arises because of the increased efficiency of fuel utilization that it implies. In a
conventional thermal electricity generating plant, about 9,000 BTU are required to generate 1 kW-hof
electricity, which corresponds to an efficiency of 38 per cent. The remaining 62 per cent of the heat
energy in the steam is discarded as waste. But when this steam can be utilized in a co-generation
scheme, the efficiency of electricity generation may rise to 80 per cent. (It is important to note that the
electricity output is limited by the utilization of steam in a co-generation scheme; that is, if electricity
output has to be increased at the same efficiency, the steam utilization must be increased
proportionately.)
Many studies have been carried out in the past on the economic and technical potential of co-generation
in Ontario. The most promising of the various co-generation alternatives, for this century, appears to
be industrial co-generation. Prompted by the potential for energy conservation in the industrial sector,
Ontario Hydro and the Ontario Ministry of Energy in December 1978 co-sponsored a seminar on
industrial co-generation.'® The seminar provided an opportunity for the representatives of Ontario
Hydro, industry, and government agencies to exchange information and discuss the economic, techni-
cal, and institutional barriers tothe implementation of co-generation.
As far as the capital costs of co-generation equipment are concerned, it is evident from various studies
that, while these costs vary widely from plant to plant, they drop steeply with an increase in the scale of
The Impact of Alternate Technologies 101
installation, up to about 15 MW. In bigger installations, the economies of scale are not as pronounced,
and the capital costs are in the $300-$400/kW range for oil- and gas-fired co-generators.!” For coal-
and low-grade-refuse-fuel-burning equipment, the capital costs are expected to be higher. The opera-
tions and maintenance costs are $10-$20/kW per year. Typical prices paid by industry for No. 2 oil,
residualoil, and natural gasin 1977 were $2.23, $1.60, and $1.75 per million BTU, respectively. The fuel
costs for co-generation in Ontario varied between $24/kW per year and $80/kW per year in 1977 with
an 80 per cent load factor. The weighted average cost was about $56/k W per year.!®
The economic potential of co-generation is analysed in detail in Volume 5 of this Report. It is shown
there that the economics of co-generation are strongly influenced by the financing environment. Pri-
vate industries expect a quick return on co-generation, since it has a lower priority than investments
that contribute directly to the increasing of output. As a result, the discount rates acceptable to the
private sector tend to be significantly higher than those used in the public sector, for example, by
Ontario Hydro. Other factors affecting the economic potential of co-generation are the cost of power
purchased from Hydro and the relative future costs of alternative fuels.
The economic analysis described in Volume 5 of this Report indicates that the additional potential for
co-generation in Ontario in both the industrial and the non-industrial sectors to the year 2000 would be
between 400 MW and 2,300 MW, depending on the assumptions about the discount rate, the price of
boiler fuel, and the price of electricity. The analysis found that, while gas-fired co-generation will be
preferred by the business sector in the 1980s because of its lower capital costs, “with financing and coal
purchasing conditions equivalent to Ontario Hydro’s, coal-fired co-generation remains cost-effective
relative to nuclear power for some time to come”.!®
In order to overcome the economic constraints, close co-operation between industry, government, and
Ontario Hydro is necessary. The importance of such co-operation is demonstrated by studies done for
the U.S. Department of Energy which estimated the co-generation potential in the U.S. to be 6,000 MW
without government action and 16,000 MW with government incentives.”°
In Ontario today, there is approximately 510 MW of installed capacity in the form of industrial co-
generation, 93 per cent of which is concentrated in plants with a capacity of more than 5 MW. Ontario’s
co-generation capacity represents about 2.5 per cent of the total capacity within the province. In
contrast to this is the situation in West Germany, where about 30 per cent of total electric generation
capacity is installed in industrial plants. The total theoretical present potential in Ontario for indus-
trial co-generation (based on estimates of steam demand) isin the order of 1,200 MW to 1,400 MW. The
technical constraints acting against utilization of this potential appear to be the specific requirements
for steam and electricity of individual plants. Industrial plants that are technically most suitable for co-
generation are those with a high load factor for both steam and electricity demand.
There are several advantages for the industrial co-generation customer in parallel operation with the
utility. Significant among these are stable frequency and voltage, increased reliability of supply, more
efficient utilization of co-generation capacity in the face of the customer’s changing need for steam and
electric power, and opportunities for the sale of excess power to the utility. Advantages to the utility
include the purchase of economy power as well as emergency power from the customer. Disadvantages
to both the customer and the utility arise from increased short-circuit currents; increased risk of
damage to the customer’s generators through faulty synchronization; the cost of additional equipment
for protection relays, synchronizing facilities, and higher short-circuit currents; and an overall in-
crease in the complexity of the operation.
An infusion into the Ontario Hydro system of many small co-generators would undoubtedly make
generation dispatch and fuel allocation more complex. But this must be weighed against the overall
benefits, mentioned earlier, for both the utility and the industrial customers, of diversification, decen-
tralization, and reductions in system reserve capacity. These reductions are considerable when an
additional industrial load is supplied by small co-generators rather than by Ontario Hydro’s large
centralized stations. Figure 4.2 of Chapter 4 showed how the reserve margin varies with the size of a
generating unit. (Throughout this discussion, it has been assumed that exchanges of power between
the utility and the industrial plants is possible in emergencies and for reasons of economy. This kind of
an arrangement is synonymous with interconnections among utilities.) As far as an increase in reli-
ability and an overall reduction in reserve requirements are concerned, the industrial customers have a
lot more to gain than Ontario Hydro has. However, Hydro’s attitude towards parallel operation with
industrial co-generators is positive despite operational complexities:
102 The Electric Power System
Some operating, protective relaying and possible short-circuit in-feed problems will result from
operation of customers’ on-site generation in parallel with Ontario Hydro’s system. These problems
have been resolved in the past by co-operation among the customer’s technical and operating staff, his
consultant and Ontario Hydro. Hydro will continue this co-operation to help encourage conservation
of finite energy resources and other benefits to the customer which result from parallel operation of
his on-site generation with the electrical supply system.”!
While the economic potential for industrial co-generation depends on the capital cost, the cost of fuel,
and the financing environment, it is the demand for steam that will determine the maximum potential.
The demand for steam in the manufacturing sector grew by 2.8 per cent per year between 1964 and
1975, a period of high economic growth. Because of the increasing role of conservation in the manufac-
turing industries, and slower industrial growth, the growth in the demand for steam up to the end of
the century is likely to be more like 2 per cent per annum. This growth rate, when applied to the existing
total potential of 1,200 MW to 1,400 MW, gives a total of about 2,000 MW by the year 2000. With about
500 MW of co-generation already existing, that means an additional potential of about 1,500 MW. This
does not include the co-generation potential at some non-industrial installations, such as universities,
hospitals, and commercial firms that operate steam plants to provide heating during the winter
months, or the self-generation of electricity by firms that have no process-steam requirements and
where one would expect to find utility-type condensing turbines. Ontario Hydro has estimated the self-
generation potential to be about 1,450 MW in 1977.”
Because of its high annual load factor, co-generation will tend to replace Ontario Hydro’s base-load
capacity requirements, that is, CANDU nuclear stations. This could cause Hydro’s load factor to deteri-
orate, with an increase in the unit cost of electricity to Ontario Hydro. However, Hydro’s position is that
“these negative effects on Hydroas the result of industrial co-generation should be viewed as only short
term. Improved load-management practices, and general load growth in the long term, offset utility
cost increases caused by the loss of some industrial customers. Furthermore, significant levels of indus-
try-owned generation plants could reduce Ontario Hydro’s capital requirements, and... the need for
new central generating capacity.””°
Biomass and Refuse-Derived Fuels
Biomass includes forest industry (timber, pulp and paper, etc.) wastes, uncommercial standing timber,
and specially planted fast-growing wood species such as hybrid poplar. Refuse-derived fuels (RDF) are
fuels obtained by processing municipal solid waste. Both biomass and RDF can be used as boiler fuel to
generate electricity.
The wood-product industry in Ontario has been using wood-waste for many years in a co-generation
mode to supply process-steam as well as electricity. The incentive to use wood-waste in this industry has
been twofold: to reduce fuel costs and to solve a waste-disposal problem. As the prices of energy from
conventional sources increase, the energy potential of wood-waste could help the lumber and pulp and
paper industries to become increasingly energy self-sufficient and to improve their competitive posi-
tions. The potential to become self-sufficient is much greater in pulp mills than in newsprint mills,
which are large consumers of electricity.
A study of the burning of wood-waste for steam and electricity production in the Township of Hearst in
northwestern Ontario has attracted considerable attention recently.”4 It was motivated originally by
problems arising in the disposal of wood-waste from the lumber mills in the Hearst area. Following a
conceptual study in 1976 by SNC Consultants, Acres Shawinigan were commissioned in 1977 by the
Ontario Ministry of the Environment to prepare a preliminary design and carry out an economic
evaluation of a wood-waste-fired steam and power plant at Hearst. Both Ontario Hydro and the Minis-
try of Energy participated actively in the study, which showed that the 117,000 tonnes of oven-dried
wood per year could supply fuel for a plant to generate 14 MW of electricity at peak and an average of
36,000 kg/hour of steam, at a total capital cost of $22.6 million (1978 dollars). According to the Ministry
of Energy, the economics of this co-generation facility are marginal under present conditions. Shell
Canada is investigating an alternative project that would use the waste to produce wood pellets as a
substitute for fossil fuels in industrial co-generators.
The hybrid poplar, developed by the Ministry of Natural Resources, has a maturing period of about 10
years compared with 30 years for existing hardwood forests. The primary aim of the fast-growing
hybrid poplar programme is to replenish Ontario’s forest resources, but the idea of cultivating these
trees for fuel also holds considerable promise. A preliminary study was undertaken for the Commission
The Impact of Alternate Technologies 103
by Morris Wayman Limited in 1978, to examine the potential of plantation wood for the generation of
electricity in Ontario.*° The study concluded that plantations of hybrid poplar in eastern Ontario could
fuel about 1,600 MW of generating capacity at a cost competitive with fossil-fuelled generation. Other
advantages cited for wood-fired generation were job creation within the province and increased reli-
ance on an indigenous and renewable resource.
Ontario Hydro undertook its own evaluation of the economics of wood-fired stations and released the
results in a report in August 1979.”° The study’s basic conclusion was that wood-fired generation is not
economically competitive with nuclear or coal-fired generation for operation at any capacity factor. An
analysis of Hydro’s cost estimates indicates that the capital and fixed operations, maintenance, and
administration (OM&A) costs of wood-fired generation are about twice those of coal-fired generation
and about 40 per cent higher than CANDU nuclear generation. The OM&A costs of wood-fired genera-
tion area very high percentage of its capital costs — from about 40 percent fora3 x 150 MW station to
as much as 120 per cent for a 24 MW single-unit station. The corresponding value for Hydro’s large
coal-fired stations is in the 25-30 per cent range. The limitations on the size of the wood-fired stations,
because of the nature of the fuel, tend to make them economically less attractive. Unless attempts are
made to reduce the capital and OM&A costs of wood-fired generation or to increase plantation yields
significantly, the rationale for developing the wood potential for electricity generation has to be other
than economic. Even if such attempts are successful, the likely role of central wood-fired generation
appears to be for intermediate-load applications. The economics of wood-fired generation may be
improved if this method is used in a specific local application in a co-generation mode. The other
technical characteristics of wood-fired generation, such as reliability, load-following capability and
part-load operation, appear to be similar to those of the fossil-fuelled units.
Ontario produces about 5.5 million tonnes of municipal refuse every year with an average heat content
of 5,000 BTU/Ib. (11.65 MJ/kg). If all of this potential were used as fuel to generate electricity, it would
produce approximately 6,000 GW-h of electricity annually, the output of a 1,000 MW station at an
annual capacity factor (ACF) of 70 per cent. However, it is unrealistic to assume that all the refuse can
be collected and all the combustible product can be recovered in preparing the fuel. A realistic estimate
of the recoverable heat potential is 60 per cent,2” which puts the maximum potential for RDF at 3,600
GW-h per year, or 600 MW at 70 per cent ACF. This corresponds to about 3.5 per cent of the demand for
electricity in Ontario in 1978.
It appears that a significant proportion of the RDF potential will be utilized ina co-generation mode to
generate electricity and provide district heating in urban centres, rather than to generate electricity
alone. This is evident from the apparent lack of success of the “watts from waste” programme involv-
ing the Ministry of the Environment, Metropolitan Toronto, and Ontario Hydro. The programme was
designed to demonstrate the feasibility of burning about 180,000 tonnes of RDF per year, mixed with
coal, in one of the units at Ontario Hydro’s Lakeview generating station. The ratio of the heat content
of coal to the heat content of RDF was expected to be about 7 to 1. However, for a variety of reasons, the
project has been suspended. A major factor was the cost of equipment for the RDF separation and
processing plant. Tenders to provide the equipment indicated that the costs would be almost twice those
expected. Final estimates of the cost of the processing plant and the facilities at the Lakeview G.S. were
in the order of $46 million.
Other alternatives were being considered to utilize Toronto’s refuse, including a scheme to burn refuse
directly in incinerators for district heating, or modifying the boilers at Ontario Hydro’s Hearn G.S.
(currently mothballed) to burn a lower-quality RDF to generate electricity as well as steam for district
heating. As a result, the Ministry of the Environment and Metropolitan Toronto are undertaking a
master-plan study using computerized models to evaluate the best alternative for using the energy
potential of Toronto refuse. The results of the study, whose cost will be shared equally by the two
participants, are expected to be available by the early spring of 1980. Another study is under way, in the
Regional Municipality of Peel, on the burning of refuse to generate electricity and provide process-
steam for a Domtar Limited plant in M ississauga.
The use of RDF for direct generation of electricity or in a co-generation mode would help to solve the
problem of municipal waste disposal and provide some diversity in fuel base by incorporating an
indigenous, renewable, and cheap source of energy. Economic considerations alone should not dictate
decisions on RDF-fuelled plants. The capital cost of the refuse-processing plants might be high, but this
must be weighed against the cost of alternate methods of waste disposal, their effects on the environ-
ment, and the savings associated with replaced fuel and the security of its supply.
104 The Electric Power System
Aleadin RDF -based electricity generation has been taken by the City of Milwaukee, Wisconsin, where
230,000 tonnes of city refuse is processed every year to fuel the boilers of Wisconsin Electric Power
Company. The electricity so generated from the RDF would supply full service to 30,000 homes.
Summary and Conclusions
The emergent technologies that are expected to have the greatest direct impact on the electric power
system of Ontario are load management, electric energy storage, co-generation, and generation from
biomass and refuse-derived fuels. Some forms of load management are already being practised by
Ontario Hydro, for example, the interruptible service to industrial customers. Ontario Hydro is under-
taking a comprehensive load-management programme to reduce primary peak loads in the 1980s and
1990s. We support Ontario Hydro load-management initiatives. Until the late 1980s, load manage-
ment may appear to be unjustifiable, because of the surplus of generating capacity faced by Ontario
Hydro. However, the cost of implementing load management is not known with any degree of cer-
tainty, and Ontario Hydro’s participation in it with the municipal utilities over the next few years will
provide the needed opportunity to assess its cost and its acceptance by the public. This will facilitate
decisions that will eventually have to be made concerning load management in comparison with other
alternatives such as peaking hydroelectric resources and storage schemes. Through the associated
hardware requirements, load management could also stimulate the province’s electronics industry.
Storage schemes serve the same objective as load management. Studies conducted by Ontario Hydro
have identified underground pumped storage as an alternative for large-scale storage of electric en-
ergy that would be viable and economic before the end of the century. The economic justification for
storage will depend on the need for peaking capacity and the availability of surplus nuclear energy
during off-peak hours. Under Ontario Hydro’s 1979 load and generation plan, the system could accept
up to 2,000 MW of storage by the late 1990s, but if the load were to grow only at 3 per cent per annum,
there might not be an economic justification for storage.
In Ontario today, there is approximately 510 MW of electrical co-generation capacity in the industrial
sector. The growth of co-generation in the industrial and non-industrial sectors will be influenced
strongly by the nature of the financing, the price of boiler fuels, and the price of electricity purchased
from Ontario Hydro. On the basis of an estimated 2 per cent annual growth in steam demand, the
additional potential for industrial co-generation by 2000 is about 1,500 MW. This does not include the
co-generation potential of some non-industrial installations. An economic analysis discussed in Vol-
ume 5 of this Report indicates that the total additional potential for co-generation to the year 2000
could be between 400 MW and 2,300 MW depending on the assumptions concerning discount rates and
the prices of fuels and purchased electricity. Any growth in co-generation will tend to replace Hydro’s
base-load requirements. Parallel operation of many co-generators with Ontario Hydro’s system will
increase the overall complexity of system operation. However, in view of Hydro’s positive attitude
towards parallel operation and the advantages of diversification, decentralization, and reduction in
system reserve capacity, co-generation should be encouraged.
The economics of wood-fired generation indicate that it cannot at present compete with coal-fired and
CANDU nuclear generation. The economics may improve if wood is used in particular local applications
in a co-generation mode. The municipal refuse in Ontario could generate about 3.5 per cent of Ontario’s
electricity demand, but a substantial part of it is likely to be used for district heating. In this respect, it
is noted that a master-plan study undertaken jointly by the Ontario Ministry of the Environment and
Metropolitan Toronto to evaluate the best way of using Toronto refuse is expected to be completed by
the spring of 1980. Utilizing the energy potential of refuse is an excellent way to solve the waste-
disposal problem. Also, it would incorporate in the system a cheap source of energy and one that is both
indigenous and renewable, and would therefore enhance the diversity and security of the fuel base.
Jair EET
Summary and Conclusions 105
vane
— >. Co see a
- ; —
: el ¢ Yate - 7
| f 7
‘. t ‘ole
7 vr re 7
« >»
‘
<
¢
&
’ @vge te?
. - =4 | ; 6 ' |
A? p ,\@ tae
, 7 ite?
Ve Prap pitas
‘é a wes
‘ <<
. ,
} a T= Cee
’ WT Sh. See Oh jae a 7
.
od pi 7 » 4
APPENDIX A
Ontario Hydro’s Generating Resources
This appendix contains a list of Ontario Hydro’s power generating resources, station-by-station for
both the East System and the West System. For each generating station, the number of units, the first
in-service dates, and the generating capacity are listed. For each of the two systems, a historical record
is given, starting in 1935, for the primary peak and energy demands and the peak generating capac-
ity. The source of this information is Ontario Hydro’s Power Resources Report No. 790201.
Definitions
Authorized Resources
These are resources that have been authorized by Ontario Hydro for design and construction. Long-
range projected resources that have not been authorized are not included.
Hydraulic Resources
Peak Resources (MW)
This refers to the peak rating of a hydraulic generating station, which is the maximum net power
available to supply system load for at least five days a week, for daily uninterrupted periods equal to:
1. Two hours for the Kast System excluding the Sir Adam Beck generating stations.
2. Twenty minutes for the Sir Adam Beck generating stations.
3. Eight hours for all West System generating stations.
Hydraulic peak and energy resources are based on monthly mean river flows. They are shown for two
alternative values:
1. “Dependable” values, attainable or exceeded 98 per cent of the time.
2. ‘‘Median” values, attainable or exceeded 50 per cent of the time.
Thermal Resources (fossil and nuclear)
Peak Resources (MW)
This refers to the peak rating of a thermal unit or generating station, which is the maximum net power
available to supply system load for a minimum of two hours a day, without exceeding specified limits of
equipment stress.
Maximum Continuous Rating (MW)
This is the expected net maximum electricity output of a unit or generating station operating continu-
ously under the designed inlet steam conditions.
Thermal Resources (combustion turbines)
Peak Resources (MW)
The peak and energy outputs of the combustion turbines vary considerably with changes in ambient
air temperature. The peak resources are based on the average of recorded daily maximum tempera-
tures for December, January, and February.
Ontario Hydro’s Generating Resources 107
Table A.l Ontario Primary Demands and Total Resources — East System
Ontario primary demand
Annual energy December dependable peak resources
Annual Fossil Combustion Total
December Average load Hydraulic steam Nuclear Purchases _ turbine dependable
Year peak(MW) GWh MW factor(%)* (MW) (MW) (MW) (MW) (MW) (MW)
1935 854.8 4,327.0 493.9 57.8 744.2 0 0 369.9 0 1,114.1
1936 927.2 4,718.7 537.2 57.9 929.1 0 0 159.3 0 1,088.4
1937 998.5 5,314.7 606.7 60.8 1,044.2 0 0 283.3 0 1,287.5
1938 1,054.3 5,303.3 605.4 57.4 1,025.9 0 0 396.5 0 1,422.7
1939 1,199.0 5,861.2 669.1 55.8 1,017.3 0 0 451.6 0 1,468.9
1940 1,266.0 7,046.8 802.2 63.4 995.2 0 0 496.5 0 1,491.7
1941 1,451.8 8,135.1 928.7 64.0 998.6 0 0 587.5 0 1,586.1
1942 1,470.1 9,054.2 1,033.6 70.3 1,030.3 0 0 648.2 0 1,678.5
1943 1,530.6 9,207.3 1,051.1 68.7 1,093.9 0 0 659.1 0 1,753.0
1944 1,612.4 9,199.3 1,047.3 65.0 1,082.8 0 0 652.2 0 1,735.1
1945 1,667.4 9,646.4 1,101.2 66.0 1,143.5 0 0 691.0 0 1,834.5
1946 1,922.0 9,791.4 1,117.7 58.2 1,173.3 0 0 708.2 0 1,881.5
1947 2,134.9 11,639.9 1,328.8 62.2 1,243.7 0 0 720.3 0 1,964.0
1948 2,245.2 12,027.9 1,369.3 61.0 1,318.9 0 0 Fin 0 2,036.6
1949 2,258.0 12,326.7. 1,407.2 62.3 1,358.5 0 0 746.0 2.0 2,106.5
1950 2,554.0 13,876.7 = 1,584.1 62.0 1,680.1 53.0 0 7464.1 0.5 2,497.7
195] 2,850.1 15,825.8 1,806.6 63.4 1,801.3 202.0 0 703.1 0.3 2,706.7
1952 3,019.3 16,910.4 1,925.1 63.8 1,960.8 444.0 0 687.1 0.3 3,092.2
1953 3,202.8 18,018.4 2,056.9 64.2 1,968.9 652.0 0 681.1 0.5 3,302.5
1954 3,389.1 18,815.6 2,147.9 63.4 2,710.9 450.0 0 681.1 0.5 3,842.5
1955 3,854.6 20,865.8 2,381.9 61.8 2,893.8 636.0 0 682.3 1.0 4,213.1
1956 4,111.8 22,870.2 2,603.6 63.3 2,922.8 616.0 0 641.2 1.3 4,181.3
1957 4,330.7 24,465.8 2,792.9 64.5 3,264.8 616.0 0 592.2 1.8 4,474.8
1958 4,644.3 25,265.6 2,844.2 62.1 4,019.8 616.0 0 593.2 1.8 5,230.8
1959 5,082.7 28,382.4 3,240.0 63.7 4322.1 616.0 0 619.2 1.8 5,559.1
1960 5,278.4 29,561.0 3,365.3 63.8 4,317.2 994.0 0 617.2 1.9 5,930.3
1961 5,480.5 30,879.2 3,525.0 64.3 4,146.2 1,372.0 0 617.5 1.6 6,137.3
1962 5,811.3 32,1368. 3,337.1 64.3 4,135.6 1,740.0 0 617.5 1.0 6,494.1
1963 6,305.1 34,517.3 3,940.3 62.5 4,437.3 2,014.0 0 617.5 1.0 7,069.8
1964 6,699.0 37,317.6 4,248.4 63.4 4,445.3 2,026.0 0 617.0 1.0 7,089.3
1965 7,344.3 40,399.4 4,611.8 62.8 4,391.4 2,526.0 0 521.3 74.0 Pot27
1966 8,028.1 44,460.9 5,075.4 63.2 4,526.4 2,588.0 0 521.5 149.0 7,784.9
1967 8,401.1 47,560.4 5,429.3 64.6 4,611.0 2,888.0 0 522.5 288.0 8,309.5
1968 9,387.5 51,771.6 5,893.9 62.8 4,931.0 3,653.0 200.0 523.6 321.0 9,628.6
1969 9,924.5 55,080.6 6,287.7 63.4 5,173.0 4,307.0 208.0 523.3 321.0 10,532.8
1970 10,638.5 59689.4 6,813.9 64.0 5,387.0 5,853.0 194.0 191.4 336.0 11,961.9
1971 10,870.4 63,396.5 7,237.0 66.6 5,879.0 5,814.0 1,204.0 193.0 343.5 13,233.5
1972 12,004.8 68,486.2 7,796.7 64.9 5,538.0 5,804.0 1,770.0 193.6 358.5 13,664.1
1973 12,858.1 73,047.3 8,338.7 64.9 5,715.0 7,334.0 2,284.0 993.9 366.0 16,692.9
1974 12,903.6 77,388.2 8,834.3 68.5 5,715.0 7,754.0 2,284.0 1,194.8 376.6 17,324.4
1975 13,974.1 79,511.5 9,076.7 65.0 5,577.0 8,321.0 2,284.0 1,195.8 384.4 17,762.2
1976 15,079.3 85,564.8 9,741.0 64.6 5,599.2 9,478.0 2,284.0 1,007.6 403.0 18,771.8
1977 14,853.9 86,964.8 9,927.5 65.5 5,709.8 10,614.0 3,764.3 9.1 445.0 20,542.2
1978 14,940.2 896143 10,2300 66.3 5,822.2 11,211.0 4,504.3 8.2 445.0 21,990.7
Note a) The annual load factor is calculated on the basis of the annual peak, which is not necessarily identical to the December peak.
108 The Electric Power System
Table A.2 Summary of Hydraulic Resources Installed and Authorized — East System
Generating
station
NIAGARA RIVER
Sir Adam Beck No. 1
Sir Adam Beck No. 2
Pumping generating station
Total
Ontario Power
Total
Toronto Power
DeCew Falls No. 1
DeCew Falls No. 2
Total
ST. LAWRENCE RIVER
Robert H. Saunders
OTTAWA RIVER
Otto Holden
Des Joachims
Chenaux
Chats Falls (Ontario half)
MADAWASKA RIVER
Mountain Chute
Barrett Chute
Barrett Chute Extension
Stewartville
Stewartville Extension
Arnprior
Calabogie
TRENT RIVER
Heeley Falls
Ranney Falls
Meyersburg
Sidney
Hagues Reach
Seymour
Frankford
Sills Island
OTONABEE RIVER
Auburn
Lakefield
MISSISSIPPI RIVER
High Falls
Galetta
RIDEAU RIVER
Merrickville
MUSKOKA RIVER
Ragged Rapids
Big Eddy
SOUTH MUSKOKA
South Falls
Trethewey Falls
Hanna Chute
—
for)
> co co CO
BRO RO PO W PO PO PO
ro Ow LS WwW WwW WwW
First
power
dates
1922-30
1905-13
0/S 1967
1906-15
0/S 1969
Q/S 1974
1904-11
0/S 1967
1943-7
1913-19
1922-6
1924
1911
1925
1909-11
1913
1926
1911-12
1928
1920
1907
1915-19
1938
1941
1916-25
1929
1926
December peak resources
Median Dependable
(kW) (kW)
1,880,000 1,880,000
105,000 28,000
Station removed from
service February 12,
1974.
155,000 155,000
776,000 702,000
247,000 225,000
420,000 418,000
116,000 110,000
96,000 84,000
167,000 165,000
42,000 42,000
130,000 130,000
64,000 64,000
103,000 102,000
78,000 78,000
4,000 3,000
11,400 11,400
8,600 8,600
5,200 5,200
3,400 3,100
3,400 3,400
3,100 3,100
2,600 2,600
1,600 1,600
1,800 1,800
1,700 1,700
2,600 2,600
800 800
900 800
7,500 7,500
7,100 7,100
4,200 4,000
1,600 1,600
1,200 1,200
Ontario Hydro’s Generating Resources
109
Table A.2 Summary of Hydraulic Resources Installed and Authorized — East System (continued)
Number First December peak resources
Generating of power —_--
station units dates Median Dependable
(kW) (kW)
BEAVER RIVER
Eugenia 3 1915-20
1 0/S 1970
Total 3,500 3,500
SEVERN RIVER
Big Chute 4 1911-19 4,300 4300
ABITIBI RIVER
Abitibi Canyon 5 1933-59 294,000 294,000
Otter Rapids 4 1961-3 179,000 177,000
MATTAGAMI RIVER ;
Little Long 2 1963 128,000 125,000
Harmon Z 1965 134,000 129,000
Kipling 2 1966 142,000 142,000
Wawaitin 4 1912-18 10,800 10,700
Sandy Falls 3 1911-16 2,700 2,600
Lower Sturgeon 2 1923 6,000 5,900
MISSISSAGI RIVER
Aubrey Falls 2 1969 158,000 158,000
George W. Rayner 2 1950 46,000 46,000
Wells 2 1970 229,000 229,000
Red Rock Falls 2 1960-61 40,000 40,000
MONTREAL RIVER
Lower Notch 2 1971 267,000 253,000
Indian Chute 2 1923-4 3,000 2,900
Hound Chute 4 1910-11 3,600 3,400
MATABITCHUAN RIVER
Matabitchuan 4 1910 10,000 10,000
SOUTH RIVER
Elliott Chute l 1929 1,200 1,200
Bingham Chute 2 1923-4 900 900
Nipissing 2 1921-4 1,600 1,600
STURGEON RIVER
Crystal Falls 4 1921 8,200 7,600
WANAPITEI RIVER
Stinson 2 1925 5,700 5,700
Coniston 3 1905-15 4,200 4,200
McVittie 2 1912 2,100 1,800
SS
seer ae hegre pera omieerageeaseane moo ep pore EE eee
110 The Electric Power System
Table A.3 Summary of Thermal Resources Installed and Authorized — East System
Generating
station
CONVENTIONAL FOSSIL-FUELLED
R.L. Hearn
Total
J.C. Keith?
Lakeview
Lambton
Nanticoke
Lennox
Wesleyvilles
COMBUSTION TURBINES®
R.L. Hearn
Lakeview
Lambton
Lennox
J.C. Keithe
Sarnia-Scott
Detweiler
A.W. Manby
Pickering A
Bruce A
Nanticoke
Bruce HWP
Pickering B
Bruce B
Wesleyville
Darlington
NUCLEAR
Nuclear Power Demonstrations
Douglas Point®
Pickering A
Pickering B
Bruce A
Darlington
Bruce B
Notes:
Number
of
units
FM SPMD WwW HPD SS Pe MH WW WwW oP oro >
ON ll
First
power
dates
1951-3
1959-61
1951-3
1961-8
1969-70
1972-8
1975-7
- 1990
1967
1967
1967
1975
1967
1965-6
1967
1965-6
1971-3
1974-6
1971
1976
1980-81
1981-3
1981
1983-5
1962
1967
1971-3
1981-3
1976-8
1987-90
1983-7
Maximum
continuous
rating
(kW)
384,000
758,000
1,142,000
254,000
2,296,000
1,980,000
3,920,000
2,140,000
990,000
22,000
206,000
2,060,000
2,064,000
2,960,000
3,524,000
3,024,000
a) Limited to 1,164,000 kW until June 1979 and to 1,171,000 kW until June 1980.
b) Unavailable until June 1, 1980.
Peak resources
(kW)
1,179,000
256,000
2,296,000
2,100,000
4,248,000
2,232,000
1,082,000
22,000
22,000
22,000
5,000
7,000
71,000
75,000
78,000!
46,000
56,000
22,000
42,000
46,000
56,000
5,000
56,000
22,000
206,000
2,060,000
2,064,000
2,960,000
3,524,000
3,024,000
c) Although Wesleyville is a part of Ontario Hydro’s committed programme, its construction has been stopped and the equipment has been stored until 1990.
d) The exact in-service dates for combustion turbines at the Bruce B, Wesleyville, and Darlington sites are not known due to recent deferrals.
e) Temporarily unavailable.
f) Limited to 58,500 kW.
g) Ontario Hydro purchases steam from Atomic Energy of Canada Ltd., which owns the nuclear part of the station.
h) Ontario Hydro purchases electricity from AECL, which owns the entire station.
Ontario Hydro’s Generating Resources
111
Table A.4 Ontario Primary Demands and Total Resources - West Systema
Ontario primary demand
December
peak(MW)
1935 52.3
1936 64.9
1937 69.5
1938 69.4
1939 73.0
1940 79.4
194] 91.4
1942 89.1
1943 100.7
1944 96.1
1945 103.9
1946 114.1
1947 50
1948 147.8
1949 185.7
1950 199.1
195] 213.0
1952 225.4
1953 240.0
1954 266.6
1955 328.6
1956 356.7
1957 406.9
1958 448.8
1959 427.9
1960 421.4
1961 422.4
1962 435.7
1963 445.5
1964 464.9
1965 474.1
1966 537.4
1967 562.7
1968 606.9
1969 630.9
1970 682.4
1971 686.8
1972 753.4
1973 752.6
1974 122.3
1975 560.8
1976 859.4
1977 845.1
1978 790.8
Notes:
a) The Kaministiquia Power Co. demand and resources are not included in the West System for the years 1935 to 1950.
Annual energy
GWh
Average
MW
68.1
Annual
load
factor(%)»
75.0
December dependable peak resources
Hydraulic
(MW)
Fossil
steam
(MW)
0
Q'S] © 2]: O C9” 2 Oooo ogc oo Co oc Soo co ooo 0 0 o.oo 2
Nuclear
(MW)
0
ooererrorocoo oo OOO CO OOOO OOOO OO DO DOO OO DODO COC COCO COC COCO COCO CO oO
Purchase
(MW)
OO Core SS O05 SocSaa 2 Sa
fom)
150.0
b) The annual load factor ib calculated on the basis of the annual peak, which is not necessarily identical to the December peak.
112 The Electric Power System
Combination Total
turbine
(MW)
75.0
0
One oo ceaocoeoocoeooteocoetoeooonmnecooecoonc Se oo Se S&S
29.0
29.0
dependable
(MW)
Table A.5 Summary of Hydraulic Resources Installed and Authorized — West System
Number First December peak resources
Generating of power
station units dates Median Dependable
(kW) (kW)
NIPIGON RIVER
Pine Portage 4 1950-54 127,200 114,800
Cameron Falls 7 1920-58 75,600 75,000
Alexander 5 1930-58 62,400 62,400
ENGLISH RIVER
Ear Falls 4 1930-48 17,800 11,100
Manitou Falls 5 1956-8 63,900 59,600
Caribou Falls 3 1958 86,100 81,400
WINNIPEG RIVER
Whitedog Falls 3 1958 68,200 61,500
KAMINISTIKWIA RIVER
Silver Falls l 1959 46,800 45,700
Kakabeka Falls 4 1906-14 23,600 18,600
AGUASABON RIVER
Aguasabon 2 1948 45,200 45,000
Table A.6 Summary of Thermal Resources Installed and Authorized — West System
Generating
station
CONVENTIONAL FOSSIL-FUELLED
Thunder Bay
Atikokan
COMBUSTION TURBINES
Thunder Bay
First
power
dates
1962-81
1984-8
1968
Maximum
continuous
rating
(kW)
391,000
412,000
Peak resources
(kW)
395,000
412,000
29,000
Ontario Hydro’s Generating Resources
113
P
= sae 2h
warms
ae —“ —— _ > = ae) -, = =
: _ Ifa k-£
; we as
ma
; rio was
<a> _ — eet? -0.e A ee
=
'
ng 7
ng wae
——— = S
: ee A "s -
Fig. B.1: p. 122
APPENDIX B
Summary of Characteristics of Conventional
Generation Technologies
This appendix outlines the characteristics — cost, reliability and performance, operating characteris-
tics, and fuel (and heavy water for CANDU plants) requirements — of the conventional electricity
generation technologies used in Ontario. These characteristics affect the choice of a generating mix
and thus the long-range expansion of the system. The factors related to the socio-environmental im-
pact of these technologies, and various characteristics of the conventional as well as the alternative
supply technologies, are dealt with in separate volumes.
CANDU Nuclear Generation
High capital costs, the use of heavy water both as moderator and primary coolant, and the use of natural
uranium asthe primary fuel are the most significant features of CANDU nuclear generation. Ontario
Hydro’s long-range plans are based on the installation of four-unit CANDU nuclear stations, ulti-
mately, to provide most of the base-load generation, except for what is provided by hydroelectric
capacity. Continued installation of new CANDU reactors carries with it the need to ensure adequate
supplies of capital, heavy water, and uranium.
In the period up to 2000, the nuclear generating units most likely to be used are 500-600 MW units
similar to those at the Pickering Generating Station (GS) and 750-850 MW units similar to those at the
Bruce GS. To reduce the transmission requirements and the associated power and energy losses, On-
tario Hydro plans to locate the new stations as close as possible to load centres but outside densely
populated areas (an example is the Pickering GS). Also, to ensure adequate supplies of natural cooling
water, these stations would probably be located on the shores of the Great Lakes and the Ottawa and St.
Lawrence rivers.
Cost
The capital cost of CANDU nuclear stations is, and is expected to continue to be, considerably higher
than the capital cost of other conventional steam-thermal plants. A study entitled “Life Cycle Costs of
Coal and Nuclear Generating Stations”, made for the Commission by Dr.S. Banerjee and Dr. L. Waver-
man in July 1978, obtained the following capital-cost components for one generating unit of a4 x 850
MW nuclear station to bein service in 1985.
1. Direct, indirect, and engineering costs (including escalation) — $785/kW
2. Interest during construction — $308/kW
3. Subtotal — $1,098/kW
4. Contingencies — $70/kW
5. Heavy water — $240/kW
6. Total capital costs — $1,4038/kW
7. Contingencies for regulatory approval delays, etc. — $24/kW
8. Imputed R&D costs — $70/kW
9. Security costs — $3/kW
10. Total cost — $1,500/kW
Items 7, 8, and 9, unique to the CANDU programme, are added to the usual capital cost. The R&D cost
component of about $70/kW is based on estimated R&D costs to date of $1.2 billion, distributed over
17,000 MW of generating capacity. In addition, an allowance of about $30 million, or $24/kW, was
made for contingencies related to regulatory and environmental processes of the sort that may delay
the construction schedule of a4 x 850 MW station. The security costs, which are negligible, are added to
show how small they are.
Figure B.1 shows Ontario Hydro’s estimates of the capital costs of CANDU, fossil-steam, and combus-
tion turbine stations to be in service in 1985. Although Ontario Hydro’s estimates of the capital costs of
an 850 MW nuclear station cannot be compared directly with the estimates of Banerjee and Waverman
because of different assumptions, it is evident that they are in the same order. Figure B.1 also indicates
Sr nnn cea EEnEnnSS SSSnSISIESnISEnEEEnanSennEEnOSEnEEE
Summary of Characteristics of Conventional Generation Technologies 115
the economies of scale, that is, the way the cost per kilowatt decreases as the size of the units is in-
creased. The economies of scale of larger units are to a certain extent offset by their lower capability
factors as well as by the higher reserve requirements associated with their greater size. This has been
discussed in Chapter 3.
The annual O&M costs for the 4 x 850 MW station were estimated in the Banerjee and Waverman
study to be: operating labour — $5.72/kW; operating materials — $3.48/kW; heavy water upkeep —
$3.26/kW; and additional security force — $2.35/k W;' totalling $14.76/kW in 1985 dollars. Figure B.2
shows Ontario Hydro’s estimates of total O&M costs for various types of generation over arange of unit
sizes and annual capacity factors (ACF).
Uranium prices began a dramatic rise early in 1973, following the quadrupling of oil prices. The spot
price for uranium in U.S. dollars increased from $16/kg in the early 1970s to about $110/kg early in
1978. In terms of electricity generation costs, this is equivalent to an increase from $0.3/M W-h to $2.2/
MW-h. However, Ontario Hydro, through forward contracts, has been able to keep its uranium cost
increases during the same period to approximately 11 per cent per annum. The fuelling cost for Picker-
ing GS in 1976 was about $1.2/MW-h. The uranium fuel costs assumed in the Banerjee and Waverman
study arein the range of $4/MW-hto$5/M W-hin 1985, which is the first year of operation.
Reliability and Performance
Ontario Hydro’s estimates of long-run average capability factors of the 500 MW and 850 MW units are
80 per cent and 77 per cent, respectively (Table B.1). These are mature values, attainable after a unit
has been in service for three or four years; the capability factors in the first year of service are expected
to be only 68 per cent and 66 per cent, respectively. Since CANDU reactors allow on-power refuelling,
their annual capability tends to be higher than that of the U.S. light-water reactors. The performance
of Pickering A has been outstanding since its commissioning in 1971; some of its units have achieved
annual capabilities of more than 90 per cent (see Table 3.4 in Chapter 3). The average lifetime capabil-
ity of Pickering A to date is about 77 per cent.
Table B.1 Ontario Hydro's 1975 Forecast of New Generating Unit Availability Indices
Fossil steam units Combustion Hydraulic
CANDU nuclear units Lignite Bituminous coal or oil turbine units units
500MW 850MW 150/200MW 300MW 500MW TS0MW
Adjusted forced outage rate
(AFOR) (%)
lst year of operation 15 15 15 15 15 17 15 0.5
2nd 12 13 13 13 12 15 15 0.5
3rd 10 12 11 1] 10 13 15 0.5
Ath 9 10 9 9 8 10 15 0.5
5th 9 10 9 9 8 10 15 0.5
Maintenance outage factor
(MOF) (%)
Ist year of operation 8 8 6 6 6 7 included included
2nd 6 6 5 5 4 5 in in
3rd 4 4 4 4 4 5 POF POF
Ath 4 4 4 4 4 5
5th 4 4 4 4 4 5
Planned outage factor
(POF) (%)
Ist year of operation 12 14 12 12 15 15 10 4
2nd 10 10 10 10 12 12 10 4
3rd 8 10 8 10 10 10 10 4
Ath 8 10 8 10 10 10 10 4
5th 8 10 8 10 10 10 10 4
Capability factor (%)
Ist year of operation 68.0 66.3 69.7 69.7 67.2 64.7 76.5 95.5
2nd 73.9 73.1 74.0 74.0 ise 70.6 76.5 95.5
3rd 79.2 7 78.3 76.5 74.4 74.0 76.5 95.5
Ath 80.1 774 80.1 78.3 79.1 76.5 76.5 95.5
5th 80.1 77.4 80.1 78.3 79.1 76.5 76.5 95.5
Source: Ontario Hydro, ‘Generation Planning Processes,’’ Submission to RCEPP, May 1976, Exh. 21.
116 The Electric Power System
Fig. B.2: p. 12
However, a recent discovery of pressure tube stretching at the four Pickering A units and the first
three of the Bruce A units will probably require each of these units to be taken out of service for one
year between 1985 and 1992, for retubing. Some of the tubes had stretched about twice as much as had
been expected. This may result in a slight reduction of the expected lifetime capability of these units.
Ontario Hydro believes that such a problem will not occur in any future reactors.
Operating Characteristics
CANDU nuclear units as now designed can be rapidly shut down and started up. However, this capabil-
ity is affected by inherent limitations in the reactor that can cause it to “poison out” (i.e., to experience
an excessive buildup of neutron-absorbing substances), and thereby be unavailable for up to 36 hours,
in some circumstances. This can occur, for instance, after a rapid shut-down from full load, if the unit is
not then quickly reloaded to a high level (typically in from 20 to 40 minutes). The likelihood of a poison-
out occurring, and the duration of a poison-out, depend, in general, on the design of the reactor and, in
particular, on the rate of shut-down and start-up. With the CANDU reactor design that is used by
Ontario Hydro, operation at full load during the daytime with a scheduled overnight shut-down is
impossible, but night-time output can be lowered to about 50 per cent of the daytime output. On week-
ends the units can be operated at fixed lower levels or shut down completely on a scheduled basis. These
CANDU units, however, are not suitable for load-following on an hour-to-hour basis.
Because of their operating characteristics and their low fuelling costs, CANDU units will be operated
essentially at base-load capacity factors for the foreseeable future. Eventually, it might be necessary
during nights and weekends either to operate them at reduced output or to “charge” energy storage
systems.
Uranium Requirements
A CANDU reactor operating at 75 per cent annual capacity factor requires 130 kg of uranium per
megawatt annually. Thus, the lifetime (30 years) requirement per megawatt of capacity is approx-
imately 4 tonnes. The present Ontario Hydro CANDU capacity of about 5,000 MW needs about 650
tonnes of uranium per year. The Canadian uranium resource estimates for 1978 are shown in Table 3.5
in Chapter 3. Most of Canada’s uranium resources lie in Ontario and Saskatchewan. A very large
proportion of Canada’s production is currently being exported to Europe and Japan. Existing export
commitments, up to 1998, amount to 62,000 tonnes of Ontario’s resources and 11,400 tonnes of Sas-
katchewan’s resources. The uranium supply and demand situation for Ontario Hydro’s programme is
discussed in more detail in Chapters 3 and 7. It is shown that Hydro’s existing contracts are sufficient
for the 30-year requirements of about 5,400 MW of nuclear capacity beyond its currently committed
programme.
Heavy-Water Requirements
The demand for heavy water by Ontario Hydro consists of central inventory demand (moderator and
primary coolant) for reactors being commissioned, and make-up demand for losses during operation.
The initial requirement is approximately 1 tonne of heavy water per megawatt of installed capacity for
a 500 MW unit, and 0.9 tonnes/MW for an 850 MW unit. The annual loss during operation is about 5
tonnes for each nuclear unit.
Heavy-water plants currently existing and under construction in Ontario include the Bruce Heavy
Water Plants (BHWP) A, B, and D. BHWP-A came into service in June 1973 and BHWP-B is expected
to be in service by March 1980. The demonstrated capacity of BHWP-A is 100.6 kg/hour. The annual
production depends on the annual capacity factor. The 1976, 1977, and 1978 capacity factors for
BHWP-A were 91 per cent, 74.5 per cent, and 80 per cent, respectively. Ontario Hydro classifies the
annual supply as “\dependable” (with a 90 per cent chance of being exceeded and corresponding to a 63
per cent annual capacity factor), “probable” (with a 50 per cent chance of being exceeded and corre-
sponding to a 73 per cent ACF), or “optimistic” (with a 10 per cent chance of being exceeded and
corresponding to an 80 per cent ACF). The annual output of BHWP-A under these three conditions will
be 550, 640, and 700 tonnes, respectively. Both BHWP-B and BHWP-D were designed for the same
capacity as BHWP-A, but in January 1979 Ontario Hydro announced the cancellation of half of
BHWP-D in the face of recent lower estimates of growth in its nuclear capacity. The total dependable
Summary of Characteristics of Conventional Generation Technologies 117
output of the three plants could support the construction of about 1,400 MW of nuclear capacity annu-
ally between 1981 and 2000, or approximately 20,000 MW beyond the currently committed
programme.
Fossil-Steam Generation
The most commonly used fossil fuels for steam-electric generation are high- or low-quality coal, resid-
ual oil, and natural gas. Ontario Hydro’s currently operating fossil-steam capacity is coal-fired, with the
exception of the residual-oil-fired 2,200 MW Lennox GS and the 1,200 MW Hearn GS, which is partly
fired by natural gas due to air quality constraints. Except for the currently committed 2 x 540 MW
Wesleyville residual-oil-fired plant, Hydro’s planned fossil-steam capacity is based on coal supplies
from the U.S. and western Canada. The current mix of Ontario Hydro’s fossil-steam resources is:
Peak capacity (MW) 1978 Energy generation (GW°h)
Coal-fired 8,965 (76%) 27,073 (87%)
Oil-fired 2,200 (19%) 1,739 (6%)
Gas-fired 600 (5%) 2,079 (7%)
Total 11,765 30,891
—--—-———————————— eee
The unit size of the coal-fired stations planned for the East System is 750 MW, and it is expected that
these stations will be located on the shores of the Great Lakes and the Ottawa and St. Lawrence rivers,
outside densely populated areas but as close to load centres as possible.
Cost
The capital cost per kilowatt of fossil-steam plants is much lower than that of CANDU nuclear plants.
The Banerjee and Waverman study obtained the following cost estimates for a 4 x 750 MW coal-fired
station coming into service in 1985.
1. Direct and indirect (including escalation) — $712/kW
2. Interest during construction — $158/kW
3. Subtotal — $870/kW
4. Subtract cost of scrubbers — $92/kW
5. Total cost — $778/kW
Items 1 and 2 are estimates for a plant with scrubbers, and, since environmental policies in Ontario do
not require the use of scrubbers, these costs are removed. Ontario Hydro’s capital-cost estimates are
shown in Figure B.1, which also indicates that the economies of scale beyond the 750 MW unit size are
marginal.
The capital costs of a gas-fired steam-thermal station are approximately 20 per cent lower than those
for a coal-fired station, because the gas-fired station has virtually no pollution control equipment, there
are no storage requirements, and the boiler furnace is smaller. An oil-fired station’s capital costs lie
somewhere between those of a coal-fired station and those of a gas-fired station, and depend on the
quality of oil used and the storage requirements. Unlike natural gas, residual oil may contain much
particulate matter and thus may require the use of electrostatic precipitators.
The above arguments also hold for comparisons of the operations and maintenance costs of the three
fossil-steam options; the costs are highest for coal, lowest for gas, and intermediate for oil. The costs for
a4 x 750 MW coal-fired station were estimated by Banerjee and Waverman to be about $9/kW per
year ($4.9/k W for labour and $4.1/kW for materials) in 1985 dollars. Ontario Hydro’s estimates, which
are comparable to the Banerjee and Waverman estimates, are shown in Figure B.2.
Ontario Hydro’s current supply of coal (about 10 million tonnes per year) is primarily from the Appala-
chian region in the eastern United States, although arrangements have been made to develop a long-
term western Canadian coal supply to supplement U.S. deliveries (see Chapter 8). The price of U.S. coal
has increased dramatically from about $10/tonne in 1970 to about $37/tonne in 1978. Ontario Hydro’s
current fuelling cost for coal-fired stations is approximately $15/MW-h. The current fuelling costs for
the Lennox oil-fired station and for Hearn’s gas-fired units are about $34/MW-h and $29/MW-h,
respectively. Hydro’s most efficient fossil-steam plants are the coal-fired Lambton GS and oil-fired
Lennox GS, with thermal efficiencies of 37.2 per cent and 37.6 per cent, respectively.
118 The Electric Power System
Reliability and Performance
If fossil-steam units are to be used for intermediate and peak loads, the planned and maintenance
outage components of unit incapability should not significantly affect unit reliability. This is because
maintenance is done during low-load periods when the output of such units may not be required.
Ontario Hydro’s estimates of the capability factors of 500 MW and 750 MW fossil-steam units (79 per
cent and 76.5 per cent, respectively) are comparable to the corresponding CANDU units, as are the
forced outage rates (Table B. 1).
Operating Characteristics
Small fossil-steam units, operating at lower temperatures and pressures, can be loaded and unloaded
rapidly and can be shut down at night without adversely affecting their reliability. Large units operat-
ing at high temperatures and pressures are not as flexible. One way to get around this is to install a
steam bypassing system to permit better control of steam temperature and pressure — the viability of
such an arrangement, however, is not well established.
Large fossil-steam units can be operated at minimum safe loadings during periods of low system
demand. These loadings are below the 50 per cent value that applies to the present CANDU nuclear
units used by Ontario Hydro. Part-load operation for extended periods, however, is quite inefficient.
Ontario Hydro’s existing fossil-steam capacity is utilized for all four modes of operation (base, interme-
diate, and peak load, as well as reserve). This pattern is expected to continue in the future, although the
base-load role of coal-fired stations will decline as new nuclear capacity is added to the system. Hydro’s
new coal-fired capacity is planned essentially for intermediate-load applications.
Fuel Requirements
A modern 1,000 MW fossil-steam plant operating at 60 per cent annual capacity factor requires ap-
proximately the equivalent of 1.8 million tonnes of bituminous coal, or 8.6 million barrels of residual oil,
or 50 billion cubic feet of natural gas per year. Of course, if the assumed capacity factor is 30 per cent,
the requirements will be half as much. In 1978, Ontario Hydro’s fossil-steam electricity generation
required 9 million tonnes of coal, 3 million barrels of residual oil, and 25 billion cubic feet of natural gas.
Ontario Hydro’s existing contracts for coal from U.S. sources are for an annual supply of 5.5 million
tonnes with the Consolidated Coal Company (expiring in 1986), 2.3 million tonnes with the Eastern
Associated Coal Corporation (expiring in 1984), 2.7 million tonnes with the U.S. Steel Corporation
(expiring in 2008), and 1.6 million tonnes with other companies (expiring in 1980). Hydro has recently
contracted for about 2.5 million tonnes per year of bituminous coal from new mines in Alberta and
British Columbia and 0.9 million tonnes per year of lignite from Saskatchewan. The lignite is for use at
the new generating units being installed at Thunder Bay, in the West System. The western Canadian
bituminous coal is of lower quality than the U.S. coal and the design of existing coal-fired stations
requires that the western Canadian coal be blended with the U.S. coal before being burned. Ontario
Hydro hassucha blending facility in operation at its8 x 500 MW Nanticoke GS on Lake Erie.
The existing supply of 5 million barrels of residual oil per year is from the Golden Eagle refinery near
Quebec City, utilizing Venezuelan and Middle Eastern crude. This contract expires in 1979 and Ontario
Hydro has signed a 15-year contract (1977 through 1991) with Petrosar Ltd. to supply 7.3 million
barrels of low-sulphur residual oil per year using western Canadian crude. The contract is renewable
for three-year periods after 1991. However, faced with reduced load forecasts, Hydro hopes to reduce
the annual supply by half. For the needs of the Hearn GS, Hydro had contracted with Consumers’ Gas
Company for 49 billion cubic feet of natural gas per year until November 1981. Again because of lower
demand, Hydro’s forecast consumption of natural gas is only 10 billion cubic feet per year until the late
1980s.
Gas Turbine Generation
Gasturbine units are available in a wide range of sizes, from under 5 MW upto 100 MW. Larger sizes of
gas turbine generating units can be devised by coupling two or more gas turbines to drive a single
generator. Ontario Hydro’s current gas turbine generating capacity is roughly 500 MW. Most of this
capacity is located at thermal generating stations, so that in addition to providing peaking capability to
the system as a whole, the units can be used to provide stand-by power for shutting down or starting up
these stations.
Summary of Characteristics of Conventional Generation Technologies 119
Gas turbines require high quality, clean, premium fuels — Ontario Hydro uses No. 2 fuel oil — because
the fuel is combusted and expanded inside the turbine and so must contain few impurities. This allows
the turbines to be located close to load centres, which means less associated transmission and voltage
control equipment.
Cost
The capital costs of gas turbine units tend to be much lower than those of fossil-steam generating units.
Ontario Hydro’s estimates (see Figure B.1) indicate that the capital costs in 1985 dollars could be
anywhere between $350/k W and $500/k W, depending on the size. Factors facilitating the low capital
costs of gas turbine units include the fact that supplies of cooling water are not needed, construction
lead times are short (one year), and nosteam generating equipment is required.
Operating labour requirements are much lower than for fossil-steam stations, but, depending on the
type of operation and the type of fuel used, maintenance costs may be quite high. For relatively
continuous operation using natural gas as the fuel, maintenance costs are low, but, for intermittent
operation using distillate oil, reliability may be poor and maintenance cost may increase. Figure B.2
shows that the operations and maintenance costs of gas turbines vary significantly with the operation.
Because premium fuels are used, the fuelling costs of gas turbines tend to be high. No. 2 fuel oil used by
Ontario Hydrocurrently costs about $40/M W-h. Natural gas works out at about $30/M W-h.
Reliability and Performance
Gas turbines suffer from the fact that their efficiency, and thus their output, decreases significantly as
the temperature of the ambient air, which is their heat sink, rises. Therefore, their capability is much
lower in summer than in winter. Their reliability is also affected by the type of operation and the type of
fuel used. The reliability of units operating relatively continuously and using natural gas is good. For
intermittent operation using distillate oil, it tends to be poor. Ontario Hydro’s estimate of a gas turbine
unit’s capability is 76.5 per cent — the forced outage rate is 15 per cent and the maintenance outage is
nearly 10 per cent. The forced outage is significantly higher than with large steam-thermal units
(Table B. 1).
Operating Characteristics
The operating flexibility of gas turbine units is not as high as that of hydroelectric units, but is higher
than that of nuclear and fossil-steam units. These units are capable of being started up and shut down
on relatively short notice, and can therefore be shut down overnight as required. However, frequent
start-ups and shut-downs increase their maintenance costs. Since the operation of gas turbines at
reduced output is very inefficient, it is preferable to run them close to rated load or shut them down.
Some units can be operated at outputs in excess of their rated load for short durations. Operating them
this way tends to increase their maintenance costs.
In Ontario Hydro’s system, gas turbines are used primarily as reserve capacity or peak capacity, or as
stand-by capacity at thermal generating stations for shutting them down and starting them up in
emergencies.
Fuel Requirements
Because of their lower thermal efficiency (in the order of 30 per cent), gas turbines require a somewhat
higher quantity of fuel, compared with fossil-steam units, to generate the same amount of electricity. A
50 MW gas turbine station operating at a 10 per cent annual capacity factor will require approximately
10,000 barrels of No. 2 oil or 50 million cubic feet of natural gas annually.
Hydroelectric Generation
Until the early 1950s, Ontario Hydro’s electric power system was based entirely on hydraulic genera-
tion, which even today supplies roughly one-third of the province’s total electricity requirements.
Hydro’s hydraulic capacity is about 6,400 MW, distributed among 70 stations varying in size from
about 1 MW to 1,400 MW. Since the operating costs of hydraulic plants are low, they are virtually
unaffected by inflation, once built. Another important characteristic of hydroelectric units is their
excellent operating flexibility. The siting of a hydroelectric generating station is limited largely by
a a a ae ec es ee ee
120 The Electric Power System
natural conditions of topography, geology, and rainfall. Some degree of flexibility is possible by the use
of water-storage developments, water diversions, tunnels, and canals.
Cost
Because the design of hydroelectric projects varies greatly from one site to another, it is not possible to
state typical capital costs for hydroelectric developments. Generally speaking, the capital costs per
kilowatt are lower for large installations, and for installations with large heads of water. Capital costs
tend to be high compared with those of large fossil-steam generating units. A study by Energy, Mines
and Resources Canada in 1978 estimated the generation capital costs of the James Bay project in
Quebec and the Gull Island project in Labrador to be $691/kW and $412/kW, respectively, in 1976
dollars. This compares with $739/kW fora CANDU nuclear station (750 MW units) and $375/kW fora
coal-fired station (750 MW units).
Most of the potential hydroelectric sites in southern Ontario have been developed and the remaining
sites are in remote areas of northern Ontario on rivers flowing into James Bay. The additional capital
costs for transmission lines, and associated power and energy losses that would occur in bringing this
power to southern Ontario loads, would increase the total unit capital cost of hydroelectric projects in
Ontario and thus make them economically less attractive. The operations and maintenance costs of
hydroelectric stations are quite low, however, compared with those of fossil-steam generation, and fuel
costs are negligible, consisting entirely of the water-rental charges levied by the government. For the
Ontario Hydro system, the average O&M and water-rental charges are approximately $1/MW-h and
$0.5/M W-h, respectively.
Reliability and Performance
Mechanically and electrically, hydraulic units are exceptionally reliable. Ontario Hydro’s estimates of
the average forced outage rate and maintenance requirements of its hydraulic units are 0.5 per cent
and 4 per cent, respectively, resulting in a capability of 95.5 per cent (Table B.1). However, the availa-
bility, or actual peak power production capability, may be adversely affected by variations in water
supply and wind, and by ice formation. The supply reliability may also be reduced by the long transmis-
sion lines.
Operating Characteristics
Hydroelectric generating units can be started up and shut down, and loaded and unloaded, quickly.
Therefore, they are most appropriate for following the daily and weekly variations in system load.
While operating them at reduced output is generally satisfactory, some units may be very inefficient at
low loadings. Therefore, in a large system such as Ontario Hydro’s, it is preferable to run them at full
output or shut them down.
Because of their good operating characteristics, hydroelectric units can be individually designed and
operated in any of the four operating modes: base load, intermediate load, peaking, or reserve. The
choice of a particular mode of operation is determined by the natural features of the site and the cost of
developing it.
Summary of Characteristics of Conventional Generation Technologies 121
Figure. B.1 Thermal Generation, Estimated Capital Cost of Nominal Capacity Coming into Service in 1985 (4-Unit Generating Stations)
3,000
Note: Estimated capital costs include net cost of commissioning
and, for nuclear units, the cost of half the initial fuel.
2,000
1,500
1985 dollars per kilowatt
1,000
500
Combustion turbine
290
900 190 1,000 1,250
Unit size (MW)
Source: ‘Generation Planning Processes”, Ontario Hydro submission to RCEPP May 1976, Exhibit 21.
Figure. B.2 Thermal Generation, Estimated Annual Operations and Maintenance Costs (4-Unit Generating Stations)
30
30
1985 dollars per kilowatt
=
200 40
Notes: Data do not include the cost of fuel consumed in the stations.
0
The 0&M costs for gas turbines vary with the annual capacity
factor. The range shown is for ACFs between 10 and 40 per cent.
CANDU nuclear, including heavy water upkeep
(ACF = 20% to 80%)
CANDU nuclear, excluding heavy water upkeep
(ACF = 20% to 80%)
600 800 1,000 1,200 1,400
Unit size (MW)
Source: “Generation Planning Processes”, Ontario Hydro submission to RCEPP, May 1976, Exhibit 21.
122 The Electric Power System
APPENDIX C
Some Technical Considerations Related to Ontario’s
Interconnections
The Ontario system and all its neighbouring systems except that of Quebec normally operate in syn-
chronism with each other with all tie lines closed. This has the advantage that mutual assistance is
immediately and atuomatically possible in an emergency. However, it carries with it the risk that, if
one utility is blacked out, it may carry its neighbours with it, as in the 1965 blackout. The National
Electric Reliability Council (NERC) is a voluntary association of Canadian and American utilities that
establishes, and monitors adherence to, system performance criteria designed to prevent this
eventuality.
When electricity utility systems operate in synchronism, power transfers can be made with considera-
ble flexibility. For instance, if it is desired to transfer power from Ontario Hydro to Detroit Edison, this
can be achieved, within broad limits imposed by transmission system capabilities, by increasing the
output of a generating station on the Ontario Hydro system, and decreasing the output of a generating
station on the Detroit Edison system, by the desired amount. However, it is not possible to maintain
perfectly smooth and steady control, and random, inadvertent transfers occur. The interconnections
must be of reasonable capacity in relation to the sizes of the systems in order to accommodate inadver-
tent exchanges as well asleave an ample margin for productive uses.
The transfer capability of an interconnection depends not only on its own capability to carry power, but
upon the point in each system to which it is joined and the capability of the system beyond that point to
transmit power to or from the interconnection. In theory, the most flexible interchange capability is
achieved by making the interconnection between the main load centres of each system since this should
best ensure that transfers are not restricted by bottlenecks in the main systems. This is seldom practi-
cable, however, and it is usually necessary to establish the shortest feasible interconnection between
suitable adjoining points on the perimeters of the grid systems. With such a configuration, the transfer
capability varies with the patterns of loads and generation and with the directions of the transfer.
Export capability is high from a perimeter interconnection when the load is low and the generation is
high in the vicinity of the interconnection. Import capability is high when the situation is the reverse.
When two systems are interconnected at two points at some distance from each other, there will in
general be a natural circulation of power around the two systems that may overload some transmission
elements of a utility and restrict their use. Circulating power around the Great Lakes, Lake Ontario
and Lake Erie in particular, is a notable example of this phenomenon. The problem of circulating power
can be mitigated by installing a phase-shifting transformer in an interconnection, as Ontario Hydro
has done at Cornwall, Windsor, and Whiteshell, Manitoba.
In Quebec, hydroelectric power is transmitted from Churchill Falls and James Bay to Quebec City and
Montreal over long and heavily loaded lines. These impose stability limitations that preclude operating
the Hydro-Québec system in synchronism with its neighbours. To transmit power to Ontario Hydro,
Hydro-Québec must disconnect a generator from its own system and connect it to the Ontario Hydro
system via one of the interconnections. For Ontario Hydro to transmit power to Hydro-Québec, the
same process is used, in reverse.
This procedure is inflexible and provides limited options for power transfers between Ontario and
Quebec. For many years the main transfers comprised firm power deliveries to Ontario Hydro that
were effected by isolating generation at Hydro-Québec’s Beauharnois plant and connecting it to the
Ontario Hydro system. This practice has been superceded by an arrangement under which Beauhar-
nois supplies the Power Authority of the State of New York in the summer months and is reconnected to
the Hydro-Québec system in the winter months.
A more flexible interconnection could be made between Ontario and Quebec by means of a converter
station, and this is being considered. A converter station functions by rectification (conversion from
alternating current to direct current) of power from the one system and its inversion (conversion from
direct current to alternating current) into the other system. While a converter station is costly, it
permits power flow in either direction to be controlled in a smooth, flexible, and rapid manner. Costs
have dropped in real terms in recent years due to the development of solid-state technology. A con-
verter station might also be used in lieu of a synchronous interconnection, although the cost might be
Some Technical Considerations Related to Ontario’s Interconnections 123
difficult to justify. Apart from considerations of cost, a converter station is superior to a synchronous
interconnection in many respects.
It is also possible to split a converter station, with the rectifier and converter sections joined by many
miles of direct-current line or cable. Direct-current lines and cables are more compact and less costly
than alternating current lines and cables. But because of the cost of the converter stations, a direct-
current link is more economic than an alternating-current link only for interconnections involving
some hundreds of miles of overhead line, or some tens of miles of cable. The conceptual plan described in
the “Report to Interprovincial Advisory Council on Energy” (IPACE) by the IPACE Networks Study
Group in October 1978 contemplates a 3,000 MW direct-current link between Winnipeg and Sudbury,
and this link might also serve to strengthen interchange capability between the East System and the
West System of Ontario Hydro.
Strengthened interconnections with the United States will probably be in the form of 500 kV alternat-
ing-current lines; however, direct-current links are also likely to be considered.
124 The Electric Power System
Notes to Chapters
Notes to Chapter Two
1. Royal Commission on Electric Power Planning. Report on the Need for Additional Bulk Power Facili-
ties in Southwestern Ontario, June 1979, and Report on the Need for Additional Bulk Power Facilities in
Eastern Ontario, July 1979.
2.In AC circuits, due to the reactance of circuit elements such as transmission lines, transformers, and
motors, the voltage and the current are generally “out of phase”, that is, they peak at different times. As
a result, power in AC circuits has two components — the active or real power, measured in watts and the
reactive power, measured in volt-amperes (reactive). Unlike real power, which is commonly referred to
as power, reactive power cannot be used to perform work and its production, except for any additional
losses it may create, does not consume energy in the form of fuel or falling water.
The presence of reactive elements in an AC power system may cause the voltage levels across the
system tochange substantially asthe power demand changes. In order to maintain satisfactory voltage
levels, provision to control reactive power must be made in the design and operation of asystem. This is
referred to as reactive power compensation.
See the Report on the Need for Additional Bulk Power Facilities in Eastern Ontario (RCEPP, July 1979)
for a discussion of voltage control.
3. Note the distinction between base load and the base-load mode of operation of a generating station.
The operating mode of a generating station is defined in terms of its capacity factor, which is the ratio
of the average power generated by a station to the station’s peak capacity. The definition of various
operating modes is rather arbitrary. Ontario Hydro defines base-load mode as annual capacity factors
(ACFs) of more than 55 per cent, intermediate-load mode as ACF's between 55 and 10 per cent, and
peak-load mode, or peaking, as ACFs of less than 10 per cent.
Notes to Chapter Three
1.S. Banerjee and L. Waverman. “Life Cycle Costs of Coal and Nuclear Generating Stations”. A study
commissioned by the RCEPP, July 1978.
2. Ontario Hydro. Cost Comparison of 4 x 750 MW Fossil-Fuelled and 4 X 850 MW CANDU Nuclear
Generating Stations
3. The loading order refers to the order in which generating units are loaded, to supply the changing
demand. Units with low fuel cost, such as nuclear and base-load hydraulic, are loaded first and supply
the base load. Those with higher fuel cost, such as oil and gas, are loaded only during peak periods or to
provide reserve. Thus, the capacity factor of a unit declines as it moves up in the loading order. Loading
order issometimes referred to as merit order or stacking order.
4. For a description of these terms and other terms used in this section, see the section entitled “Availa-
bility and Security in the Generation Subsystem” in Chapter 4.
5. Sierra Club of Ontario. “Planning Electric Power for Ontario”. Submission to the RCEPP, Septem-
ber 1978. RCEPP Exhibit 369, p. 61.
6. While Bruce is essentially an 850 MW reactor, its electrical generating capacity is only 750 MW,
because part of the steam is supplied to the heavy-water plants.
7. R.L. Scott. “Outages at Light-Water-Reactor Power Plants: A Review of 1973-1977 Experience”,
Nuclear Safety, vol. 20, no. 2, March-April 1979, p. 211.
8. RCEPP transcript vol. 800, p. 44952.
9. 1,000 BTU/Ib is equivalent to 2.33 MJ/kg in S.I. units. Thus the heat contents of the western
Canadian and U.S. bituminous coals are 25.63 MJ/kg and 30.3 MJ/kg, respectively.
10. RCEPP interrogatory no. 21-26.
11. RCEPP transcript vol. 39, p. 4894.
12. Sierra Club of Ontario. Op. cit., pp. 80-8 1.
13. Jbid.
14. Ontario Hydro. “Generation Planning Process”. Submission to the RCEPP, May 1976. RCEPP
Exhibit 21.
15. The operating limitations of the CANDU reactors at low capacity factors refer to their limited
capability in daily load-following. In isolation, a CANDU unit, or any unit for that matter, can be run at
NotestoChapters 125
any low annual capacity factor simply by running it at full output over a given time and shutting it
down for the rest of the year.
16. RCEPP transcript vol. 236, pp. 37380-37384.
17.Ontario Hydro. Op. cit., p. 50.
18. RCEPP transcript vol. 235, pp. 87279-3728 1.
19. Ontario Hydro. Op. cit.
20. Ontario Hydro. “Total Electric Power System”. Submission to the RCEPP, October 1978. RCEPP
Exhibit 375, p. 12.
Notes to Chapter Four
1. The difference between the availability and the capability factor of a unit may be explained as
follows. Availability takes into account the capability as well as the factors external to the unit, such as
fuel and water shortages and strikes. Therefore, availability is usually less than or equal to the capabil-
ity of a unit. In the absence of any external restraints, availability and capability may be used
interchangeably.
2. Report on the Questionnaire on Generating Capacity Reliability Evaluation, prepared by the Power
System Reliability Committee, Canadian Electrical Association, March 1975. The results of this survey
were also reported by R. Billinton in his submission to the Ontario Select Committee on Hydro Affairs
on March 21, 1976.
3. The discussion of the LOLP methodologies of the three utilities is based on the information obtained
through the 1975 survey indicated in Note 2. While we are aware of some subsequent changes in
Ontario Hydro’s methodology, we do not know of any recent developments in reliability evaluation in
Hydro-Québec or in Manitoba Hydro. The changes iii Ontario Hydro’s methodology will be discussed in
subsequent sections.
4. Ontario Hydro. “System Expansion Program Reassessment Study”. Final Report, February 1979.
Also see interim reports 2 and 5 published in November 1978.
Notes to Chapter Five
1. Ontario Hydro. ‘‘Hydroscope”’, July 27, 1979, vol. 16, no. 15.
2. Ontario Hydro. ‘‘Total Electric Power System”. Submission to the RCEPP, October 1978. RCEPP
Exhibit 375. Since Ontario Hydro’s submission in October 1978, the accident at the Three Mile Island
nuclear reactor has opened up the possibility of firm exports to Pennsylvania. Ontario Hydro and
General Public Utilities (GPU), the owner of the TMI plant, are negotiating a contract under which
Ontario Hydro would export 1,000 MW of firm power to GPU over the period 1985-90. The power would
flow from the coal-fired Nanticoke Generating Station via an underwater cable (200-400 kV) to Erie,
Pennsylvania. This would interconnect Ontario Hydro with a new power pool, that is, the Pennsy]-
vania-New Jersey-Maryland pool. It is believed that the capital cost of the cable would be recovered ina
couple of years.
3. [bid.
4. RCEPP. Report on the Need for Additional Bulk Power Facilities in Eastern Ontario. July 1979, pp. 96-
fi:
5. U.S. Department of Energy, and Energy, Mines and Resources Canada. ‘“‘Canada/United States-
Electricity Exchanges”. May 1979.
6. Energy, Mines and Resources Canada. “An Energy Strategy for Canada: Policies for Self-Reliance”.
Ottawa, 1976.
7.““An Evaluation of Strengthened Interprovincial Interconnections of Electric Power Systems”. Re-
port to Interprovincial Advisory Council on Energy by IPACE Networks Study Group. October 1978.
8. Ontario Hydro Statistical Yearbook, 1977.
Notes to Chapter Seven
1. Ontario Hydro. “Load Forecasting”. Submission to the RCEPP, May 1976. RCEPP Exhibit 19, table
4,
2. Ontario Hydro. “Survey on Power System Reliability: Viewpoint of Large Users”. Report No. PMA
76-5, April 1977. See also the discussion on the costs and benefits of reliability in Chapter 4. Some
customers, such as hospitals, computer installations, and certain processing industries, have their own
stand-by generation to ensure essential services and to prevent adverse effects in vulnerable processes.
3. RCEPP transcript vol. 236, p. 3738 1.
ee eee ee ee oe ee ee rep le ee eae ee
126 The Electric Power System
4, Ontario Hydro. “Generation Planning Processes”. Submission to the RCEPP, May 1976. RCEPP
Exhibit 21, pp. 65-7.
5. A New Public Policy Direction for Ontario Hydro. Final report of the Select Committee of the Legisla-
ture Investigating Ontario Hydro. June 1976.
6. For an example of the methodology used earlier, see ‘Planning of the Ontario Hydro East System”.
Ontario Hydro Report 573 SP, June 1976.
7. The 1,100 MW and 600 MW of co-generation capacity refers to the type of in-plant generation by
which steam produced for process purposes is passed through a turbo-generator to produce electricity
as a by-product. Other types are co-generation potential, requiring investment in new steam facilities,
and self-generation of power by firms or institutions that have no process-steam requirement. The
total technical potential in 1985 of the three types of in-plant generation is estimated by Ontario Hydro
to be 1,125 MW, 765 MW, and 1,730 MW respectively.
8. Ontario Hydro. “1979 Review of Generation Expansion Program”. March 1979.
9. Ontario Hydro. “System Expansion Program Reassessment Study”. Final report. February 1979, p.
17.
10. Ontario Hydro. Op. cit., p. 9.11.
11. Ibid., p. 5.4.
12. RCEPP. Report on the Need for Additional Bulk Power Facilities in Southwestern Ontario, June 1979,
p. 95.
13. Jbid., p.85.
14. Ontario Hydro. Op. cit., p. 138.14.
15. Ontario Hydro. “Transmission Planning Processes”. Submission to the RCEPP. RCEPP Exhibit 22,
June 1976.
16. For a detailed discussion of the issues related to bulk power transmission planning in southwestern
and eastern Ontario, see the RCEPP’s Report on the Need for Additional Bulk Power Facilities in
Southwestern Ontario, June 1979, and Report on the Need for Additional Bulk Power Facilities in East-
ern Ontario, July 1979.
17. Ibid.
18. N. Hingorani. “The Re-emergence of DC in Modern Power Systems”, EPRI Journal, June 1978.
Notes to Chapter Eight
1. Ontario Ministry of Energy. Ontario Energy Review. June 1979, p. 31.
2. Ontario Hydro. “Electricity Costing and Pricing Study”. October 1976. The OKB submitted its Report
to the Minister of Energy on Principles of Electricity Costing and Pricing for Ontario Hydro on December
20, 1979. The OEB concluded that “the concept of time-differentiated rates is consistent with the
fairness objective” (p. viii) and recommended that “the concept of time-differentiated rates be intro-
duced at both the bulk power and retail levels” (p. 36).
3. RCEPP. Interim Report on Nuclear Power in Ontario. September 1978, p. 20.
4, Ontario Hydro. *1979 Review of Generation Expansion Program”. March 1979.
5. RCEPP. Report on the Need for Additional Bulk Power Facilities in Eastern Ontario. July 1979, pp. 90-
Oo:
6. Cost Study on Intermediate Storage in Industry for Ontario Hydro. Vol. 1. Consultee Ltd. and H.H.
Bush and Associates Ltd. February 1978. RCEPP Exhibit 375-5.
7.Ontario Hydro. “The Role for Load Managementin Ontario”. July 1978. RCEPP Exhibit 375-2.
8. Ibid.
9. Ontario Hydro. “1979 Review of Generation Expansion Program”. March 1979.
10. Ontario Hydro. ‘Preliminary Study of Energy Storage Alternatives”. January 1975. RCEPP Ex-
hibit 198.
11. Ontario Hydro Underground Pumped Storage Study”. Acres Consulting Services Limited, Niag-
ara Falls, Ontario, January 1976.
12. Ibid., p. 6.
13. Jbid.
14. Letter to Dr. W. W. Stevenson of the RCEPP from G. F. McIntyre of Ontario Hydro’s Resources
Planning Department, September 4, 1979.
15. Ontario Hydro. “Total Electric Power System”. Submission to the RCEPP, October 1978. RCEPP
Exhibit 375, p. 11.
16. The Economics of Industrial Co-Generation of Electricity. Proceedings of a seminar co-sponsored by
the Ontario Ministry of Energy and Ontario Hydro. December 1978.
NotestoChapters 127
17. [bid., paper by Donald D. Dick.
18. Comments by Ontario Hydroon the Middleton Associates November 1977 report entitled “Alterna-
tives to Ontario Hydro’s Generation Program”. June 1978.
19. Vol. 5of this Report.
20. See the paper by A. Juchymenkoin the seminar proceedings referred toin note 16.
21.See the paper by A. Gusen inthe seminar proceedings referred to in note 16.
22.See the paper by D.A. Drink walter inthe seminar proceedings referred toin note 16.
23. Ontario Hydro. “Development of Industrial Co-Generation in Ontario”. Report ECD-78-8. Novem-
ber 1978, p. 11.
24. See “Cost of Production of Electricity with Wood By-Products: Hearst Study” in the seminar pro-
ceedings referred to in note 16.
25.'"Wood-Fired Electricity Generation in Eastern Ontario”. Study prepared for the RCEPP by Morris
Wayman Limited. July 1978.
26. Ontario Hydro. “Wood as a Fuel for Electric Power Generation in Ontario Hydro’s System”. Report
79207, August 1979.
27. Ontario Hydro. ‘‘Generation-Technical”. Submission to the RCEPP, March 1976. RCEPP Exhibit 2,
p. 2.2-37.
Notes to Appendix B
1. Itis recognized that the study’s estimates of the additional security force are too high. However, since
they do not affect the nuclear-coal comparison to any significant degree, no attempt was made to revise them.
128 The Electric Power System
+ :
PRY i
a i, y i?
caper
roy
ye
an
Wy try
*
ae h a4
’ yl ve
by ed,
4 ie
2
sb
"
Wh
‘
‘,
t
i
ae
i
M
Live Music
Archive
Librivox
Free Audio
Metropolitan Museum
Cleveland
Museum of Art
Internet
Arcade
Console Living Room
Open Library
American
Libraries
TV News
Understanding
9/11